Corrosion Behavior of Carbon Steel in CO2

Republic of Iraq Ministry of Higher Education & Scientific Research University of Baghdad College of Engineering Chemical Engineering Department

Corrosion Behavior of Carbon Steel in CO2 - Containing Oilfield Produced Water A Thesis Submitted to the Chemical Engineering Department of the College of Engineering of the University of Baghdad In Partial Fulfillment of the Requirements for The Degree of Doctor of Philosophy in Chemical Engineering

By Khalid Hamid Rashid Abdul-Khalik B.Sc. & M.Sc. in Chemical Engineering College of Engineering/ University of Baghdad

May – 2014

Dedication

To

My Wife Nebaal, Girls Mays,Mina and Rusul, BrothersMajid andHazim & SisterKhulood And every one that helped me during my study, I present this work with my deep respect and Love….. Khalid

Acknowledgment

Acknowledgment

Acknowledgment First of all I thank Allah who gives me patience and strength to continue…….. Secondly, I would like to express my sincere thanks and deep gratitude to my supervisor Prof. Dr. Aprael S. Yaro for his valuable support and I would like to thank his for all the discussions and the useful suggestions, and also for the patience that he always had with me. I would like also to express my grateful admiration to Assist. Prof. Dr. Hasan F. Makki, head of the Chemical Engineering Department and Assist. Prof. Dr. Basma A. Abdul Majeed . My great appreciation to all the staff of the Department of Chemical Engineering / University of Baghdad for providing the Facilities for this work. I would also like to thank all my teachers in the Chemical Engineering Department and my colleagues in the laboratory for their supporting to me. I would like to express my very deep respect to my lovely family for their encouragement, which gave me so much hope during my study. I wish to express my respect and appreciation to my aunt Nazhat Salman Abdul-Khalik and my brothers and sister, especially Dr. Hazim. My warmly thanks to all my friends, especially Assist. Lecturer Salman Hussein Abbas for their support. Finally, thanks to all people who assisted me. Khalid I

Summary

Summary

Corrosion Behavior of Carbon Steel in CO2 - Containing Oilfield Produced Water By

Khalid Hamid R. Abdul-Khalik Supervised by

Prof. Dr. Aprael S. Yaro Department of Chemical Engineering College of Engineering-University of Baghdad Carbon dioxide is present in water as a dissolved gas under the high pressures common in underground oil and gas reservoirs. In the dissolved state it forms carbonic acid. The primary material of construction for pipelines in the oil and gas industry is mild steel, because of its price, strength and availability. However, carbon steel corrodes in the presence of carbonic and organic acids such as acetic acid (HAc). It is therefore important to investigate the conditions in which HAc causes corrosion damage to mild steel. The extent of HAc/CO2 corrosion depends on many other variables such as: temperature, CO2 partial pressure, pH, flow regime, etc. The corrosion rates of API X65 mild steel alloy have been studied by three different techniques: i. Weight Loss Technique ii. Potentiodynamic Polarization Technique iii. Characterization of the Corroded Surface Techniques

i. Weight Loss Technique: A series of experiments were performed to study the effect of simulated brines solutions on the corrosion rate of mild steel with and

II

Summary

Summary

without acetic acid. The corrosion rates of mild steel were found to be similar in simulated brines solutions and 3.5 wt % NaCl solutions. The corrosion experiments were planned to form a second-order mathematical expression using Full Factorial Experimental Design (FFED): a- Four variables experiments (influence of temperature, solution pH, acetic acid (HAc) concentration and speed of rotation). b- Three variables experiments (influence of temperature, solution pH and speed of rotation). The results of this investigation are summarized as follows: The second-order polynomial regression analysis of the objective function (corrosion rate), using Full Factorial Experimental Design (FFED) via STATISTICA software, gave two mathematical expressions for four and three variables experiments. Arrhenius Equation and Transition State Equation were used to evaluate the activation parameters: Activation Energy (Ea), Enthalpy of Activation (△H*) and Entropy of Activation (△S*). The values of average Equilibrium Constants (K*) were also calculated at each value of average Gibbs Free Energy Change (ΔG), to determine the spontaneous of corrosion reaction. The corrosion rate of mild steel in presence and absence of acetic acid were increased with increasing of temperature, acetic acid (HAc) concentration and speed of rotation, but decreased with increasing of pH of solution. Multi-variable regression analysis of objective function (corrosion rate) in presence and absence of acetic acid in weight loss technique as a function of experimental variables (temperature, solution pH, acetic acid (HAc) concentration and speed of rotation), yielded two suggested mathematical expressions.

III

Summary

Summary

ii. Potentiodynamic Polarization Technique: In this investigation a theoretical model Equation proposed by Korobove and Medvedeva Korobove and Medvedeva, [2000] was used to analyze the shape of polarization curves. The values of polarization resistance (Rp) were also obtained, these values were increased with decreasing of temperature and speed of rotation in absence and presence of the protective film formation. The polarization resistance values in absence of acetic acid are larger than the polarization resistance values in presence of acetic acid, due to the formation of the protective film. The values of mass transfer correction factor (λ) were also obtained, these values will approach unity at low overpotential and it decreases as overpotential increases in presence and absence of the protective film formation. Generally, in absence of acetic acid, the values of (λ) are adjacent to each other and almost unite value compare with presence of acetic acid at different temperatures and speeds of rotation due to the protective film formation as diffusion barrier is accelerated by measures that restrict the transport of reaction products from the surface. The limiting diffusion currents of hydrogen in CO2 saturated, 3.5 wt% NaCl solutions under turbulent conditions in presence and absence of the acetic acid has been correlated.

iii. Characterization of the Corroded Surface Techniques: The effect of presence and absence of the acetic acid (HAc) on the CO2 corrosion of grade API X65 mild steel alloy was investigated at optimum conditions in weight loss technique (45.4 °C, pH 4.8, 2178.5 ppm HAc and 1296.6 rpm) and in absence of acetic acid (68.7 °C, pH 7.9 and 1425.8 rpm) by using analyses of protective film thickness, porosity, roughness,

Vickers

micro-hardness

IV

(VMH),

scanning

electron

Summary

Summary

microscopy (SEM), computerized metallurgical optical microscopy technique (CMOMT) and X-ray diffraction (XRD). In presence of acetic acid, a porous layer (Fe3C cementite/FeCO3 siderite layer) was formed. In absence of acetic acid, a fairly dense layer ferrous carbonate (FeCO3 / siderite layer) was formed. In absence of acetic acid, the roughness and hardness of protective film were greater than that of film formation in presence of acetic acid.

V

Contents

Contents

Contents Subject

Pages

Acknowledgments……………………………………………..... Summary………………………………………………………... Contents…………………………………………………………. Nomenclature…………………………………………………....

I II VI X

Chapter One: Introduction 1.1 Overview…………………………………………………….. 1.2 Objectives of This Research………………………………….

1 5

Chapter Two: CO2 Corrosion of Mild Steel 2.1 Introduction to Corrosion……………………………………. 2.2 Overview CO2 Corrosion……………………………………. 2.3 Corrosion Attack in Oil and Gas Pipeline…………………… 2.4 Chemistry of Carbon Dioxide Saturated Media…………….. 2.5 Carbon Dioxide Corrosion…………………………………... 2.6 Mechanisms of Carbon Dioxide Corrosion………………….. 2.7 Acetic Acid Corrosion……………………………………….. 2.8 Forms of CO2 Corrosion…………………………………….. 2.8.1 Uniform Attacks………………………………………… 2.8.2 Localized Attacks……………………………………….. 2.9 Environmental Factors Influencing CO2 Corrosion…………. 2.9.1 Effect of Temperature…………………………………... 2.9.2 Effect of pH……………………………………………... 2.9.3 Effect of Speed of Rotation……………………………... 2.9.4 Effect of Acetic Acid (HAc) Concentration…………….. 2.10 Corrosion Product Film Formation………………………… 2.10.1 Iron Carbide (Fe3C)……………………………………. 2.10.2 Iron (II) Carbonate (FeCO3)…………………………… 2.11 Electrochemical Measurements……………………………. 2.11.1 Polarization……………………………………………. (i) Activation Polarization (ηA)…………………………... (ii) Concentration Polarization (ηC)………………………. (iii) Resistance Polarization (ηR)………………………….. (iv) Combined Polarization (ηT)…………………………... 2.11.2 Mass Transfer………………………………………….. 2.11.3 Corrosion Rate Measurements…………………………. (i) Linear Polarization Technique………………………… (ii) Tafel Extrapolation Technique………………………... (iii) Effect of Concentration Polarization on the Determination of Corrosion Rates from Polarization Measurements……………………………………………… 2.12 Research Methodology of Building an Experimental Design

VI

7 7 9 10 11 13 16 18 18 19 20 20 27 28 30 31 31 32 33 33 35 37 39 40 41 44 44 45 46 48

Contents

Contents

and Optimization for CO2 Corrosion………………............... 2.13 Regression Analysis………………………………………... 2.14 Modeling and Theoretical Analysis of CO2 Corrosion System……………………………………………………....

51 56

Chapter Three: Experimental Work 3.1 Materials……………………………………………………... 3.1.1 Corrosive Solution……………………………………… 3.1.2 Carbon Dioxide CO2 Gas……………………………….. 3.1.3 Chemicals……………………………………………….. 3.1.4 Working Electrodes……………………………………... 3.1.5 Cleaning Materials……………………………………… 3.2 Equipment and Accessories…………………………………. 3.2.1 Equipment………………………………………………. 3.2.2 Accessories……………………………………………… 3.3 Weight Loss Measurement………………………………….. 3.4 Design of Experimental Polarization Holder………………... (i) Teflon Shaft Description…………………………………... (ii) Procedure………………………………………………….. 3.5 Potentiodynamic Polarization Technique…………………… 3.5.1 System Description……………………………………... 3.5.2 Preparation……………………………………………… 3.5.3 Open Circuit Potential Measurement…………………… 3.5.4 Potentiodynamic Polarization Measurements…………... 3.6 Tests for Protective and Non Protective Films……………… 3.6.1 Protective Film Thickness Test…………………………. 3.6.2 X-Ray Diffraction………………………………………. 3.6.3 Roughness Test…………………………………………. 3.6.4 Hardness Test…………………………………………… 3.6.5 Microstructure Examination……………………………..

60 60 61 61 62 63 63 63 64 64 66 66 67 68 68 69 69 70 77 77 78 79 80 81

Chapter Four: Results and Discussion 4.1 Weight Loss Measurements…………………………………. 4.1.1 Experiments in a Simulated Brine Solution…………….. 4.1.2 Experimental Strategy in NaCl Solutions………………. 4.1.3 No Protective Film Formation………………………….. 4.1.3.1 Study Area………………………………………. 4.1.3.2 Experimental Response………………………….... 4.1.3.3 Used Matrix……………………………………….. 4.1.3.4 Statistical Treatment of Data……………………… 4.1.3.5 Validity of the Model…………………………....... 4.1.3.6 Graphic Analysis of the Model………………........ (i) Evolution of Corrosion Rate as a Function of the Temperature and the pH……………………………….

VII

82 82 82 82 83 84 84 86 89 93 93

Contents

Contents

(ii) Evolution of Corrosion Rate as a Function of the pH and HAc Acid Concentration……………………… (iii) Evolution of Corrosion Rate as a Function of the Temperature and Speed of Rotation…………………... (iv) Evolution of Corrosion Rate as a Function of the HAc Acid Concentration and Speed of Rotation……… 4.1.4 In Presence of Protective Film………………………….. 4.1.4.1 Study Area………………………………………... 4.1.4.2 Experimental Response…………………………... 4.1.4.3 Used Matrix………………………………………. 4.1.4.4 Statistical Treatment of Data……………………... 4.1.4.5 Validity of the Model…………………………..... 4.1.4.6 Graphic Analysis of the Model…………………... (i) Evolution of Corrosion Rate as a Function of the Temperature and the pH………………………………. (ii) Evolution of Corrosion Rate as a Function of the Temperature and Speed of Rotation…………………... (iii) Evolution of Corrosion Rate as a Function of the pH and Speed of Rotation…………………................... 4.1.5 The Effect of Experimental Variables…………………... 4.1.5.1 Effect of Temperature……………………………. 4.1.5.2 Effect of pH………………………………………. 4.1.5.3 Effect of Acetic Acid Concentration……………... 4.1.5.4 Effect of Speed of Rotation………………………. 4.1.6 Combined Influence of Temperature, pH, Acetic Acid Concentration and Speed of Rotation on the Corrosion Rate……………………………………………………….. 4.1.6.1 No Protective Film Formation…………………… 4.1.6.2 In Presence of Protective Film………………….... 4.2 Electrochemical Results……………………………………... 4.2.1 Open Circuit Potential Measurements (OCP)…………... 4.2.2 API X65 Mild Steel Potentiodynamic Polarization Curves…………………………………………………… 4.2.2.1 API X65 Potentiodynamic Polarization Curves (No Protective Film Formation)…………………………. 4.2.2.2 API X65 Potentiodynamic Polarization Curves (Under Protective Film Formation)……………………… 4.2.3 Parameters Estimated from API X65 Polarization Curves………………………………………………….. 4.2.3.1 Corrosion Potentials and Corrosion Current Densities (Tafel Extrapolation Method)……………….. 4.2.3.2 Corrosion Potentials and Corrosion Current Densities (McLaughlin Method)………………………..

VIII

95 97 99 101 102 103 103 104 107 111 111 112 114 116 116 127 129 132 135 136 138 140 140 142 144 150 157 157 165

Contents

Contents

4.2.3.3 Corrosion Process Kinetic Parameters (β-Model).. 4.2.3.4 API X65 Mild Steel Mass Transfer Correction Factors…………………………………………………… (i) API X65 Mass Transfer Correction Factors (No Protective Film Formation)………………….. (ii) API X65 Mass Transfer Correction Factors (Under Protective Film Formation)……………….. 4.2.3.5 API X65 Mild Steel Polarization Resistances……. (i) API X65 Mild Steel Polarization Resistances (No Protective Film Formation)………………….. (ii) API X65 Mild Steel Polarization Resistances (Under Protective Film Formation)………………. 4.2.3.6 Evaluation of Corrosion Parameters……………… 4.2.3.7 Effect of Environmental Conditions on Hydrogen Limiting Current………………………………….. (i) Effect of Re……………………………………… (ii) Effect of Sc……………………………………… 4.3 Characterization of the Corroded Surface…………………… 4.3.1 API X65 Mild Steel Protective Film Thickness Test…… 4.3.2 API X65 Mild Steel Roughness Test…………………… 4.3.3 API X65 Mild Steel X-Ray Diffraction………………… 4.3.4 API X65 Mild Steel Hardness Test……………………...

167 169 170 177 184 187 190 193 197 198 202 215 215 215 221 223

Chapter Five: Conclusions & Recommendations for Further Work 5.1 Conclusions………………………………………………….. 5.1.1 Weight Loss Technique…………………………………. 5.1.2 Potentiodynamic Polarization Technique……………….. 5.1.3Characterization of the Corroded Surface Techniques….. 5.2 Recommendation for Further Work………………………….

References Appendices

IX

225 225 227 229 230 231

Nomenclature

Nomenclature

Nomenclature Symbol

Units m2

A

Definition Surface Area

A

Pre-Exponential Factor (Frequency Factor) Constant in Arrhenius-type Equation for

J.mol-1

Constant in Arrhenius-type Equation for

J.mol-1

Free Term (Parameter) of the Mathematical Model

[-]

Linear Terms

[-]

Quadratic Terms

[-]

Interaction Terms

[-]

ba

= (k x 1) Vector of the First-Order Regression Coefficients = (k x k) Symmetric Matrix, whose Main Diagonal Elements are the Pure Quadratic Coefficients, while Off-Diagonal Elements are One-Half Mixed Quadratic Coefficients Anodic Tafel Slope

bc

Cathodic Tafel Slope

CA Cb

Acetic Acid (HAc) Concentration Bulk Aqueous Concentration of H+ ions Constant “Near-Zero” Concentration of H+ underneath mackinawite Film at Steel Surface, set to 1x 10-7 kmol/m3 Diffusion Coefficient of Reacting Ion Outside Diameter of Specimen Electrode Potential Standard Electrode Potential Over Potential Activation Energy Corrosion Potential Equilibrium Electrode Potential Faradays Constant Film Thickness Gibbs Free Energy Change Enthalpy of Activation

b

B

Cs D dc E E°

△E Ea Ecorr Eeq F Ft △G △H* h icorr io ilim

gmd

Plankُ s Constant Corrosion Current Density Exchange Current Density Limiting Current Density Mass Transfer (Diffusion) Limiting Current Density for H+ ion reduction

X

[-]

[-]

V.decade-1 V.decade-1 ppm mol.m-3 [-] mol.m-3 m2/s m V V V J.mol-1 V V 96500 C.mol-1 µm J.mol-1 J.mol-1 6.626 x 10-34 J.s.molecule-1 A.m-2 A.m-2 A.m-2 A.m-2

Nomenclature Symbol IR k k K*

Nomenclature

Definition Ohmic Potential Drop Conductivity of the Electrolyte Solution Number of the Input Factors Equilibrium Constant Equilibrium Constant for Dissociation of (H2CO3) Equilibrium Constant for Dissociation of (HCO3-) Kinetic Constant in Ferrous Carbonate Precipitation Rate Equation Solubility Constant for Dissolution of CO2 Equilibrium Hydration Constant for CO2, Khyd= = 2.58x10-3 Forward Reaction Rate for CO2 Hydration Reaction, Backward Reaction Rate of H2CO3 Dehydration Reaction,

Ksp(FeCO3) km L n, z N

N pH R R Rf Rp Rs △S*

Solubility Product Constant for Ferrous Carbonate Mass Transfer Coefficient Equilibrium Constant for Dissociation of (H2O) Length of a Specimen Number of Electron (Electrical Valence of Ions) Avogadr’s Number Number of Experiments Hydrogen Ion Concentration (-log [H+]) Correlation Coefficient Universal Gas Constant Resistance of Films Polarization Resistance Bulk Solution Resistance Entropy of Activation

Sb2 Sr2 T

Variance of Coefficient

t W1 W2

Time Weight of a Specimen before Weight Loss Weight of a Specimen after Weight Loss Weight of a Specimen with the Film Mass of Film per unit Area Weight of Mild Steel Reacted Weight of a Specimen after Film Removal = (k x 1) Vector of the Independent Variables Canonical Independent Variables (Factors)

WM x {xi}

Units V Ω .cm-1 [-] [-] mol/m3 mol/m3 -1

mol-1 s-1 mol/m3/bar [-] s-1 s-1 (mol/m3)2 m/s (mol/m3)2 m [-] 6.022 x 1023 molecule.mol-1 [-] [-] [-] 8.314 J.mol-1.K-1 Ω Ω.cm2 Ω J.mol-1.K-1 [-] [-]

Experimental Error Variance Temperature of Solution

XI

°C s g g g g.m-2 g.m-2 g [-] [-]

Nomenclature Symbol xi Xi

Nomenclature Units

Definition Coded Variables (Factors) Natural Variables

[-] [-]

Central Value

[-] [-]

̂ Y

gmd (g/m2.day) gmd (g/m2.day) gmd (g/m2.day)

Estimated Response Estimated Natural Response Measured Response

Greek Symbols Symbol ρ ρ βa βc α β ηA ηR ηC {θi} Å µ v

γ λ

η ηct ηmt |δi | |̅| ω Å



Definition Specific Resistance (Resistivity) Density of Medium Density of a 100% FeCO3 Film Anodic Tafel Slope Cathodic Tafel Slope Diffusion Layer Thickness Transfer Coefficient (Symmetry Factor) = (icorr/ilim) Activation Polarization Resistance Polarization Concentration Polarization Eigenvalues ( Canonical Coefficients) Angstrom Dynamic Viscosity of Medium Turbulent Diffusion Coefficient Kinematic Viscosity IR-Drop Correction Factor Mass-Transport Correction Factor Over Potential Charge-Transfer Over Potential Mass-Transport Over Potential Experimental Error (Corresponding Residual) Absolute Percentage Error Mean Absolute Percentage Error

Units Ω.cm kg/m3 kg/m3 V.decade-1 V.decade-1 m [-] [-] V V V [-] Å kg/m.s (Pa.s) m2/s m2/s [-] [-] V V V [-] % %

rpm

Speed of Rotation Film Porosity Angstrom No. Degree of Freedom

[-] Å [-]

Abbreviations Symbol SCE OCP

Units

Definition Saturated Calomel Electrode Open Circuit Potential

[-] [-]

XII

Nomenclature LPR ppm rpm C.R C.Rp lim h.p. HAc XRD Exp. WL SEM gmd ANOVA

Nomenclature

Linear Polarization Resistance Part per Million Revolution per Minute Corrosion Rate Predicted Corrosion Rate Limiting Current Density Horse Power Acetic Acid X-ray Diffraction Experimental Weight Loss Scanning Electron Microscopy Gram per Meter Square per Day Analysis of Variance

[-] [-] [-] [-] [-] [-] [-] [-] [-] [-] [-] [-] [-] [-]

Subscripts Symbol a c corr app ads A C R k

Definition Anodic Cathodic Corrosion Applied Adsorbed Activation Concentration Resistance Number of the Input Factors

Units [-] [-] [-] [-] [-] [-] [-] [-] [-]

Superscripts Symbol ct mt T A C k

Definition Charge-Transfer Mass-Transport Total Anodic Cathodic Number of the Input Factors

Units [-] [-] [-] [-] [-] [-]

Dimensionless Groups  ρud    μ 

Re

Reynolds Number 

Sc

Schmidt Number 

Sh

Sherwood Number 

[-]

 μ    ρD 

[-]

 k m .d    D 

[-]

XIII

Chapter One Introduction

Chapter One

Introduction

1.1 Overview:

C

arbon steels are generally used for the petroleum industry for transportation of crude oils and gasses over long distances from

their sources to ultimate consumers and from these to different destination of the applications. So that corrosion problems exist in the oil industry at every stage of production from initial extraction to refining and storage Migahed, [2005]. Carbon dioxide (CO2) corrosion of carbon steel pipelines and equipment in the oil and gas industry has been given much attention in recent years because of an increased tendency to inject CO2 into oil wells to reduce the viscosity of oil and increase its production. Despite the fact that carbon steel has low resistance to CO2 environments, it is widely used in the petroleum industry mainly due to economical reasons Jiang et al., [2006]. An important fact is that when CO2 dissolves in water, carbonic acid (H2CO3) is produced with the characteristic that is more aggressive than hydrochloric acid at the same pH Zhang et al., [2007]. What makes feasible the use of carbon steels is the natural precipitation of protective iron carbonate (FeCO3) Farelas et al., [2010]. Internal corrosion in flowlines and pipelines is influenced by temperature, CO2 and HAc content, water chemistry, flow velocity, oil or water wetting and composition and surface condition of the steel. A small change in one of these parameters can change the corrosion rate considerably. In presence of CO2, the corrosion rate can be reduced substantially under conditions when corrosion product, iron carbonate (FeCO3) can precipitate on the steel surface and form a dense and protective corrosion product film. This occurs more easily at high temperature or high pH in the water phase. When corrosion products are not deposited on the steel surface, very high corrosion rates of several millimeters per year can occur. To avoid the consequences of corrosion, 1

Chapter One

Introduction

process parameters should always be controlled within safe operating limits under no protective film formation. To do so, corrosion rates at various values of the parameters are to be predicted to set the critical values of every parameter; and then the process should be operated below these critical values. This can be achieved based on experimental measurement or statistical modeling. The models available in literature for corrosion prediction are most likely empirical models based on flow or electrochemical data Mysara et al., [2010]. Gaseous carbon dioxide and water vapor are commonly coproduced components in most of oil and gas fields. Although dry CO2 gas is non-corrosive in nature to steels at moderate temperatures, its hydration in the condensing water forms an aggressive environment that may damage facilities through the so-called CO2 corrosion or “sweet corrosion”. The corrosiveness of such environment depends on pH, temperature and CO2 partial pressure, beyond the inherent properties of the steel and the hydrodynamic conditions prevailing. The hydration of CO2 is a slow chemical reaction that can become the controlling step for the overall CO2 corrosion rate George and Nesic, [2007]. CO2 corrosion has been extensively investigated over many decades and a considerable knowledge is now available in literature, particularly in which relates to corrosion-induced failure of oil and gas pipelines Gulbrandsen and Bilkova, [2006] and Nesic, [2007]. CO2 corrosion is a complex electrochemical phenomenon in which simultaneous coupled processes occur. Although factors that play key roles in CO2 corrosion has been identified, a satisfactory mechanistic explanation of their mutual interactions has not yet been given Schmitt and Horstemeier, [2006]. CO2 corrosion is often associated with the presence of some other acidic gases or volatile short chain carboxylic acids usually co-produced 2

Chapter One

Introduction

with unprocessed hydrocarbons. Acetic acid appears to be one of the most prevalent organic acids found in oil and gas reservoirs with high concentrations up to thousands of ppm in the produced aqueous phase. The effect of HAc on the corrosion rate of carbon steel in CO2-containing oilfield produced water has been extensively investigated over a wide range of conditions during the last few years. To date, however, the fundamental role of acetic acid in CO2 corrosion of carbon steel has been the subject of apparent controversies, particularly in which concerns the eventual electroactive participation of this compound in the cathodic mechanism. The effect of HAc on the overall kinetic behavior of carbon steel and corrosion scaling was addressed by means of electrochemical measurements. Whether HAc acts as a specific cathodic reactant or just as a proton source Kermani and Morshed, [2003]. CO2 corrosion in the presence and absence of acetic acid (HAc) is recognized as a major cause of premature failure and protection and mitigation from corrosion of mild steel pipelines in the oil and gas industry. CO2 corrosion in the presence of HAc has been the subject of numerous studies since 1983 and particularly in the late 1990s. It is found that HAc can significantly increase the CO2 corrosion rate. The CO2 corrosion rate is known to be flow dependent, where flow increases corrosion rates by increasing the mass transfer of corrosion species and/or by damaging the protective film on the steel surface Nafday, [2004]. CO2 is present as a dissolved gas in the oilfield produced water/brine that accompanies oil production at high pressures common in underground oil and gas reservoirs. In the dissolved state it forms carbonic acid. The brine is largely a NaCl solution, but it also contains other metal ions such as Na+,K+,Ca+2, Mg+2, Cl-, SO4-2 and organic acids the most common of which is HAc. Premature failure results in millions of dollars in property damage worldwide besides lost production and bodily injury. It is caused 3

Chapter One

Introduction

by the presence of a complex variety of flow regimes, multiphase flow conditions and the presence of organic acids the most common of which is HAc. However carbon steel has a tendency to corrode in the presence of CO2 and organic acids such as acetic acid. It is therefore important to investigate the conditions in which HAc causes corrosion damage under no protective film formation and the conditions without HAc made protection under protective film formation Nesic and Nafday, [2005] and Song et al., [2004]. In CO2 corrosion, iron carbonate (FeCO3) film is the chief corrosion product formed and is formed through the reaction between carbonic acid, source of carbonate ions, (CO3-2) and iron (Fe+2) released through corrosion of the pipeline. FeCO3 forms on the wall of the pipe if the product of ferrous ion concentration (Fe+2) and carbonate ion concentration (CO3-2) exceeds the solubility product limit. The film is known to be protective and the corrosion rate drops once the film starts growing. Although iron carbonate film formation mechanisms and kinetics have been extensively studied, it is not known how protective the film will be in the presence HAc. Moreover it is not known if the film failure (if any) is a result of a lower system pH or the result of interaction between corrosion products and HAc. Thus it becomes imperative to understand how FeCO3 precipitation is affected in the presence of HAc, as also by the pH, temperature and ionic strength of the solution Sun et al., [2003]. A problem constantly faced by chemical engineers is that of determining optimum operating conditions for a process. Toward this end, effects of a number of variables must be experimentally, evaluated. Here it requires finding conditions under no protective film formation of temperature, pH, concentration of acetic acid (HAc) and speed of rotation, and finding conditions under protective film formation 4

Chapter One

Introduction

of temperature, pH and speed of rotation used which will minimize the corrosion rate of carbon steel in CO2 saturated, 3.5 wt % NaCl solution prepared experimentally. A well known technique or tool for optimizing will be used which has been successfully applied to industrial problems, it is called the statistical Full Factorial Experimental Design (FFED) or Factorial method. It minimizes the number of experiments required for determining an optimum point by taking the steepest slope path (in a number of sequential steps) to the optimum point. The present work is a step in the direction of understanding the corrosion of carbon steel by rotating a cylindrically shaped test specimen in NaCl solution saturated with CO2 at different conditions in presence and absence of acetic acid. Keeping in mind that working with tube flow requires a bulky and cumbersome apparatus, while rotating electrodes give a compact and easily handled apparatus, and are therefore preferred as reported by Egil and Katerina, [2006]. Finally:

1.2 Objectives of This Research: The objective of this study is to investigate the effect of free HAc which is known to be a source of hydrogen ions and to lead to an increase in mild steel corrosion rates. The scope of study based on the objectives can be simplified as follows: 1. To investigate the effect of temperature of CO2 saturated, 3.5 wt % NaCl solutions in the range of (40-75 °C) in presence and absence of acetic acid (HAc) on corrosion rate of API X65 mild steel. 2. To investigate the effect of speed of rotation of the working electrode in CO2 saturated, 3.5 wt % NaCl solution in the range of (1000-1500 rpm) in presence and absence of acetic acid (HAc) on corrosion rate of API X65 mild steel. 5

Chapter One

Introduction

3. To investigate the effect of pH of CO2 saturated, 3.5 wt % NaCl solution in the range of (7.5- 8.5) in absence of acetic acid (HAc) on corrosion rate of API X65 mild steel. 4. To investigate the effect of presence of acetic acid (HAc) on the kinetics and corrosion parameters of the cathodic reaction compared with its absence.

6

Chapter Two CO2 Corrosion of Mild Steel

Chapter Two

CO2 Corrosion of Mild Steel

2.1 Introduction to Corrosion:

C

orrosion is a general term for a reaction between a metal and its environment that causes the metal to breakdown. While there are

many types of corrosion, they all involve either a chemical reaction or an electrochemical reaction. In chemical reactions, chemicals in the environment react with the metal to create different chemicals. Thus, atoms or molecules of the metal combine with other atoms or molecules that contact the metal to form different corrosion products. FeCO3 or Fe3C is an example of this type of corrosion. In electrochemical corrosion, the environment around the metal results in the creation of an electrical current, which is simply a flow of electrons. The metal corrodes by giving up electrons to create the electrical flow. The oil field environment is filled with metal pipes and other components that often exposed to chemicals that can cause corrosion, especially when the metal and chemicals are in a solution such as downhole fluids. The pumper must understand how to reduce corrosive damage to the metal in wells, flow lines, tank batteries and equipment. There are three general types of corrosion of concern in the oil field. These involve three chemicals of concern and electrochemical corrosion. The types of corrosion include Villamizar et al., [2007]: I. Carbon dioxide (sweet corrosion) II. Hydrogen sulfide (sour corrosion) III. Oxygen corrosion (oxidation)

2.2 Overview CO2 Corrosion: Carbon dioxide (CO2) corrosion is one the most studied form of corrosion in oil and gas industry. This is generally due to the fact that the crude oil and natural gas from the oil reservoir / gas well usually contains 7

Chapter Two


some level of CO2. The major concern with CO2 corrosion in oil and gas industry is that CO2 corrosion can cause failure on the equipment especially the main down hole tubing and transmission pipelines and thus can disrupt the oil/gas production. Figure 2.1 shows the model of CO2 corrosion of a crude oil pipeline made of mild steel. The study of CO2 corrosion rate and FeCO3 film formation are essential to enhance the understanding and modeling the kinetics of FeCO3 precipitation process. The presence of CO2 in solution would initiate the CO2 corrosion process. It would produce a weak carbonic acid (H2CO3) which is corrosive to carbon steel or low alloy steel and it is presented by Equation (2.1) below Abd El-Lateef et al., [2012]: (



)

(

)

The reaction process will continue with three cathodic reactions (reduction) and one anodic reaction (oxidation). The cathodic reactions in CO2 solutions are: 1. Reduction of carbonic acid into bicarbonate ions. 

(

)

(

)

(

)

2. Reduction of bicarbonate ions into carbonate ions. 

3. Reduction of hydrogen ions. 

There are some main factors that can affect the severity of CO2 Corrosion Rate (C.R). The first factor is CO2 partial pressure: higher partial pressure of CO2, C.R will be higher. The second is temperature: higher temperature, higher C.R while the third is pH will result in higher C.R. Flow velocity also can affect the severity of CO2 corrosion: consequence higher velocity is higher C.R Chen et al., [2000].

8

Chapter Two


 ( ) FeCO3 H+

H2

 Steel pipe wall

Fig. 2.1 Model CO2 Corrosion of Crude Oil Pipeline Made of Mild Steel Bin Muhammad, [2013].

2.3 Corrosion Attack in Oil and Gas Pipeline: The presence of carbon dioxide (CO2) and free water can cause severe corrosion problems in oil and gas pipelines. Internal corrosion in wells and pipelines is influenced by temperature, CO2 content, water chemistry, flow velocity, oil or water wetting and composition and surface condition of the steel. A small change in one of these parameters can change the corrosion rate considerably, due to changes in the properties of the thin layer of corrosion products that accumulates on the steel surface Nyborg, [2004]. When corrosion products are not deposited on the steel surface, very high corrosion rates of several millimeters per year can occur. The corrosion rate can be reduced substantially under conditions where iron carbonate (FeCO3) can precipitate on the steel surface and form a dense and protective corrosion product film. This occurs more easily at high temperature or high pH in the water phase Koteeswaran, [2010]. Localized corrosion with very high corrosion rates can occur when the corrosion product film does not give sufficient protection, and this is the 9

Chapter Two


most feared type of corrosion attack in oil and gas pipelines. The line had been in operation for several years without problems, but changes in the well composition over time led to more aggressive conditions, resulting in unacceptably high corrosion rates. In order to control the corrosion in pipelines, it is important to understand the underlying corrosion mechanisms and be able to predict whether localized corrosion will be initiated and how it can be prevented Koteeswaran, [2010].

2.4 Chemistry of Carbon Dioxide Saturated Media: Carbon dioxide gas corrosion significance in the aqueous media is associated to its solubility where the aggressive species are produced with different concentrations. Carbon dioxide solubility is basically dependant on temperature (T) and on the partial pressure (PCO2). In conditions where the partial pressure of carbon dioxide is 1 bar, the solubility, expressed with mole fraction (y) in a range of temperature from 273 to 353 K as: R ln(y) = a + b T-1+ c ln (T) + d T

… (2.5)

The factors a, b, c and d are assigned to specific pressures but for 1-bar-CO2-saturated media, they are expressed as: a = -1327.8 J/K.mol, b =72611.6 J/mol, c = 180 J/K.mol and d = -0.009 J/K2.mol. R is the universal gas constant and equals 8.314 J/mol. K. The above relation is plotted in Figure 2.2 showing the decreasing mole fractions of the dissolved carbon dioxide with the higher temperature.

10

Chapter Two


Fig. 2.2 Mole Fraction Variations of Carbon Dioxide with temperature from 273 to 353 K in 1 bar CO2 Saturated Aqueous Media Faysal, [2011].

2.5 Carbon Dioxide Corrosion: CO2 corrosion, or “sweet corrosion,” of carbon steel is not a new problem. It was first recorded in the U.S. oil and gas industry in the 1940s, followed by several studies since then Garsany et al., [2002]. Dry CO2 gas by itself is not corrosive at the temperatures encountered within oil and gas production. It needs to be dissolved in an aqueous phase to promote an electrochemical reaction between steel and the contacting aqueous phase. CO2 is soluble in water and brines. However, it should be noted that it has a similar solubility in both the gaseous and liquid hydrocarbon phases. Thus, for a mixed-phase system, the presence of hydrocarbon phase may provide a ready reservoir of CO2 to partition into the aqueous phase.CO2 is usually present in produced fluids. The basic CO2 corrosion reaction mechanisms have been well understood and accepted by many researchers through the work done over the past few decades. Research into the corrosion mechanisms of carbon dioxide and its effects on mild steel under varying conditions of pressure, temperature, pH and oil-water fractioning has been done by Xiao and Nesic, [2005] and Nesic et al., [2001]. They have proposed 11

Chapter Two


models to predict carbon dioxide corrosion of mild steel based on the results of their work. The major chemical reactions include CO2 gas dissolves into water to form carbonic acid through hydration by water as shown in ( ) (

)

(

)



(

)

(

)

The carbonic acid then dissociates to form bicarbonate which itself can further dissociate 

(

)



(

)

Accordingly, the reduction of the undissociated acid molecule (H2CO3) occurs after it is absorbed onto the metal surface. This is the ratedetermining step of the process, so the corrosion rate of the metal surface is directly related to the concentration of the undissociated acid in solution. Two possible cathodic reactions occur in the corrosion process  

(

)

(

)

(

)

while a single anodic reaction occurs 

Despite more than three decades of intense research, it is still not known which of the two reactions (2.10) or (2.11) actually occur on the metal surface. Hence, the net cathodic current was assumed to be the sum of the currents of the two cathodic reactions. It has been suggested that the direct reduction of carbonic acid becomes important at higher pH George and Nesic, [2007]. Whether or not the direct reduction of carbonic acid (Equation 2.10) actually occurs on the metal surface is debated since it could be 12

Chapter Two


argued that carbonic acid would dissociate into a hydrogen ion faster than it could diffuse to the surface of the steel. If carbonic acid dissociates in the boundary layer, then it would only act as an additional source of hydrogen ions and the only cathodic reaction in the corrosion process is (Equation 2.11). It has also been shown that the solubility of iron carbonate in salt water decreases with an increase in system temperature. This iron carbonate precipitate may form a protective or a non protective film depending on the solution composition, pressure, temperature and the environmental conditions of the system George et al., [2004]. The overall reaction in the corrosion process can then be written as 

(

)

The effect of flow on corrosion when no protective films are present is through increased mass transport of the corrosion species to the metal surface. When the mass transport of the species is fast enough to support the electrochemical process, limiting currents result. On the other hand, accumulation, saturation, and precipitation can result at the metal surface if the transport of the species away from the surface is limited. If the corrosion process is under charge transfer (activation control) or chemical reaction control, then changes in the flow will have no effect on the corrosion rate since mass transfer is not the limiting step.

2.6 Mechanisms of Carbon Dioxide Corrosion: Corrosion of pipeline steels in CO2-saturated media was reported in many studies as a complex phenomenon requiring great research efforts to investigate its mechanisms and the associated determining steps. Many views were proposed, but however, they were for describing the corrosion reactions in specific cases and they were not also widely recognized 13

Chapter Two


Kermani and Morshed, [2003]. Basically, CO2 corrosion is an electrochemical process of a multistep nature between the corrosive species resulting from CO2 dissolution and the dissolvable phases in the steel. A great effort has been devoted to determine the governing anodic and cathodic reactions and to determine the key species involved. In literature, there is an agreement on three simultaneously occurring processes which are the anodic dissolution of iron to ferrous ions, cathodic reduction of carbon carrying species producing hydrogen, and the formation of effectively stable solid corrosion products; summarized as: 

(

)

These processes were employed to provide an understanding on the possible governing mechanism(s) as performed in Nesic, [2007] and Zhang et al., [2006] and Paolinelli, [2008]. Most models proposed take primarily into account the chemical equilibrium among the reducible species resulting from carbon dioxide dissolution as Uhlig, [2011]: 

( )

(

)

(

)

(

)



  

(

)

(

)

(

)

(

)

(

)

Except with slight differences on reporting the importance of these species with respect to their ability to drive the corrosion reactions upon temperature or pressure variations, it was almost widely agreed that the cathodic reactions govern the overall process. 14

Chapter Two


In a compiled synopsis by Burke, it was indicated that hydrogen evolution in the deoxygenated CO2-saturated environments is the controlling process in the corrosion reactions Burke, [1984]. In that early effort published, it was also reported from other 3 works that the limiting cathodic currents were higher in CO2-saturated solutions of adjusted pH levels of 4, than those in CO2-free solutions of the same pH level. Dugstad confirmed the dependence of the uniform corrosion kinetics on hydrogen evolution where the process involves the heterogeneous hydration of chemisorbed CO2 as the rate determining step Dugstad, [2006]. However, this process was found to be necessarily temperature dependent in conditions where adherent corrosion product formations are not facilitated; at temperatures above 60

o

C the

permeability and/or solubility of iron carbonate (FeCO3) formed more effectively controlled the corrosion reactions. Hydrogen evolution is preceded by multistep reduction reactions of carbonic acid, bicarbonate, and hydrogen protons at extents depending fundamentally on pH levels making the reactions to occur near to interface or upon adsorption. In one electrochemical chain, hydrogen protons resulting from carbonic acid dissociation diffuse to the steel surface where they get reduced as: (

( (



)



)

(

)

(

)

)



(

(

)

)

(

)

(

)

(

)

(

)

Alternatively, carbon dioxide was proposed to get adsorbed and then hydrated producing adsorbed carbonic acid molecules where they get reduced directly, as proposed in Zhang et al., [2006], or serving as sources for hydrogen protons as: 15

Chapter Two

(

)

(

)


(

(

)

(

)

(

)



(

)

)

(

)

)



(

)

(

)

(





(

(

)

)

At higher pH levels, carbonic acid can dissociate further producing bicarbonate species driving the cathodic reactions as proposed by George and Nesic George and Nesic, [2004] as:  (

(

)



)

(

)

(

)

The anodic reaction is widely represented by the direct dissolution of iron to ferrous ions as: 

(

)

The corrosion products involving for example iron carbonate (Ksp(FeCO3) = 10.54 at 25 oC Tong et al., [2008] form by the direct combination with carbonate or via reaction intermediates as Fajardo et al., [2013] :  

(

)

) 

(

(

)

(

)

(

)

2.7 Acetic Acid Corrosion: When a gaseous phase of HAc is present in multiphase pipelines (or from the dissociation of an acetate compound to form HAc), in addition to carbon dioxide, dissolves into the aqueous solution. The HAc then dissociates into hydrogen and acetate: 

(

16

)

Chapter Two


Since HAc is a stronger acid than carbonic acid (pKa 4.76 vs 6.35 at 25 °C), it is the main source of hydrogen ions when the two acid concentrations are similar. The acetate ions form iron acetate upon reaction with iron Tran et al., [2013]: 

(

) (

)

Moreover, iron acetate’s solubility is much higher than iron carbonateُ s, so protective film formation by iron acetate does not readily occur. Without formation of a stable protective film, the corrosion rates of the steel can remain at a high value. Hedges and McVeigh, [1999] published results on acetateُ s role in CO2 corrosion. Experiments using both HAc and sodium acetate as a source of acetate ions in various media (3% NaCl and two synthetic oilfield brines) were performed using rotating cylinder electrodes. Both sources of acetate ions were shown to increase the corrosion rate, but acetic acid decreased the pH while sodium acetate increased it. The increased corrosion rates were attributed to the forming of thinner iron carbonate films since acetate ions have the ability to form iron acetate and transport iron away from the steel surface. However, no attempt was made to quantify the thickness or morphology of the films formed in their experiments. Gunaltun, [2000] reported that very few systematic studies have been performed in the laboratory. Little or no information exists about the basic effect of HAc on the anodic and cathodic reactions. Dougherty, [2004] reported a mild increase in the cathodic reaction in the presence of HAc although their results were not fully conclusive. The work of Crolet and Bonis, [2005] suggests that the presence of HAc inhibits the anodic (iron dissolution) reaction.

17

Chapter Two


Garsany et al., [2002] performed work using voltammetry to study the effect of acetate ions on the rates and mechanisms of corrosion using a rotating disc electrode (RDE) on film-free surfaces. Their voltammograms show two waves, which are attributed to hydrogen ion and HAc reduction on the steel surface. They argue that since HAc dissociation can occur very quickly it is not possible to distinguish the reduction of hydrogen ions from direct HAc reduction at the electrode surface. Joosten et al., [2002] performed additional experiments using acetic acid, synthetic seawater, and an oil phase in glass cells. They found that acetic acid increased the corrosion rate by decreasing the pH, but the system could be inhibited very effectively (below 400 ppm HAc). Effective inhibition has also been reported by Hadges et al. The most important feature of the Joosten et al., work is the presence of pitting on the 13%Cr electrode. This material would be the likely material chosen for service if X-65 was found unacceptable. However, no pits were found on a super 13%Cr material under the test conditions (600 ppm HAc and 95 °C). Sun et al., [2003] recently published work using potentiodynamic sweeps to the study the effect of HAc on the cathodic and anodic reactions using a rotating cylinder electrode (RCE). Their work suggests HAc acts solely as an additional source of hydrogen ions and is not reduced at the surface.

2.8 Forms of CO2 Corrosion: In CO2-containing aqueous media, corrosion attacks may take several forms that strongly depend on the inherent characteristics of the metal, the environmental conditions prevailing, and the kinetics of the

18

Chapter Two


interfacial corrosion processes. There are mainly two forms of corrosion often encountered in the field, namely uniform and localized corrosion.

2.8.1 Uniform Attacks: Uniform corrosion, known also as general corrosion, takes place across the entire exposed surface. Nevertheless, the damage caused by this form of corrosion is predictable. Uniform corrosion is also by far the most widely studied form of attack in oil and gas industry. A number of predictive models have been devoted to this form of corrosion over the last two decades. A comparison of the respective performances and accuracies of these models has been reported in literature Wang et al., [2002]. Recently, Nyborg has published a detailed overview on the major models used in the field Nyborg, [2002]. The history and mathematical schemes has also been summarized Kapusta et al., [2004].

2.8.2 Localized Attacks: Localized corrosion is known to be more severe and more difficult to both predict and control than uniform corrosion due respectively to its latent incubation, quick propagation and stochastic nature. This form of attack implies that only discrete parts of the metal surface are attacked. Localized corrosion often results in deep penetration of the metal, thus inducing an impairment of its characteristics. In practice, localized corrosion is at the root of most corrosion failures of facilities encountered in oil and gas fields. This is particularly observed at low fluid velocities for flowing systems or in stagnant conditions. A rational explanation would be that a pre-initiated localized attack is prone to grow in such conditions due to local electrochemical concentration gradients between the attack and the outer surrounding surface. Due to its severity, localized CO2 corrosion attacks have been given an increasing attention.

19

Chapter Two


Modelling approaches were also used to tackle the stochastic nature of this phenomenon Xiao and Nesic, [2005]. Under corrosion scaling condition, particular localized attacks known as mesa attacks can develop. These attacks are particularly observed at high fluid velocities. For this reason, the phenomenon of mesa attacks is generally categorized as flow-induced or flow-assisted localized corrosion. The prerequisite for the formation of mesa attacks is that partially protective corrosion scales must form first. Mesa attacks are typically characterized by deep and often flat-bottomed cavities. The mechanisms for initiation and growth of mesa corrosion attacks have not been given a satisfactory explanation until the work of Nyborg and Dugstad, [2003]. The mechanism suggested by Nyborg is sketched in Figure 2.3. Some other efforts have been made to better understand this phenomenon Schmitt et al., [2000]. Since mesa attacks are localized, their occurrence is also unpredictable.

Fig. 2.3 Mechanism for Initiation and Growth of Mesa Attacks According to Nyborg, [1998].

2.9 Environmental Factors Influencing CO2 Corrosion: Considering the increasing need for long-term performance of metallic structures in oil and gas fields over a wide size scale, the effective use of any steel must be based on the understanding of its interaction with environmental conditions prevailing, along with its intrinsic properties. Having briefly outlined CO2 corrosion mechanisms, 20

Chapter Two


one can anticipate that there are many key factors that affect CO2 corrosion in oil and gas industry.

2.9.1 Effect of Temperature: Temperature increases the rate of almost all-chemical reactions. An increase in temperature of a corroding system has four main effects: 1. The rate of chemical reaction is increased. 2. The solubility of gases in solution is decreased. 3. The solubility of some of the reaction products may change, resulting in a different corrosion reaction products; and 4. Viscosity is decreased, and many thermal differences will results in increased circulation Uhlig, [2011]. The magnitude of the temperature effect for CO2 corrosion reaction has been studied by a number of workers Amri, [2009] and Ibraheem et al., [2012], most often in presence and absence of acetic acid. In many of these investigations, the number of temperature points obtained is too small to yield a very accurate value of activation energy. So they put their data into an Arrhenius type Equation:  E  Corrosion Rate C.R.  A Exp   a   RT

(2.36)

Where: A = Modified frequency factor (pre-exponential factor) Ea = Activation energy (J/mole) R = Gas constant (8.314 J/mole. K) T = Absolute temperature (K). C.R. = Corrosion rate (g/m2.day). By taking the logarithms of the Equation (2.36) then: log C.R.  log A 

Ea 2.303 R T

… (2.36a)

21

Chapter Two


So that Equation (2.36a) is a straight line relation, log (C.R.) gives a straight line when plotted vs. (1/T) with a slope of (-Ea /2.303 R). Many thermally activated processes behave in a similar way and an Arrhenius plot enables us to determine the activation energy Shreir et al. (2), [2000]. Temperature changes have the greatest effect when the ratedetermining step is the activation process. In general if the corrosion rates are doubled for a certain increase in temperature, activation process may be increased by (10-100 times), depending on the magnitude of the activation energy, bearing in mind the importance of the rate determining process and because of the complex situation in corrosion processes of having two electrode processes Shreir et al. (1), [2000]. Activation parameters for some systems can be calculated from an Arrhenius-type plot Equation (2.36) and from transition state theory; this theory shares certain similarities with collision theory. Reaction is postulated as occurring when molecules collide with or encounter each other. A transition state complex of relatively high energy is formed; the complex is then decays to products. This is pictured schematically in Figure 2.4 for general reaction;

O  J  P V

... (2.37) OJ*

O+J B

P+V

Fig. 2.4 Formation of Activated Complex Peter and Julio, [2009]. 22

Chapter Two


Where OJ* is the activated complex or transition state complex. The reaction coordinate simply depicts the progress of reaction from reactants towards products via the transition state complex. Transition state complex is postulated to be in equilibrium with reactant: … (2.38)

O  J  OJ *

OJ   *

K

*

… (2.39)

OJ 

Where K* is equilibrium constant. Further the rate of reaction is postulated to depend only on the rate of decay of OJ* to products; K3 K1 O  J  OJ   P V

… (2.40)

K2

The rate reaction (r) is then;

 

r  K 3 OJ *

... (2.41)

From Equation (2.39) the concentration of OJ* can be obtained;

And

r  K 3 K * OJ 

... (2.42)

r  K OJ 

... (2.43)

Where K = K3 K*

… (2.44)

Statistical mechanical consideration predicts that the state complex will decay with a rate constant; K3 

RT Nh

… (2.45)

Where: R: the gas constant (8.314 J.K-1.mol-1) N: Avogadr`s number (6.022 x 1023 molecule.mol-1) h: Plank’s constant (6.626 x10-34 J.s.molecule-1) So Equation (2.44) will be;

23

Chapter Two


 RT  * K  K  Nh 

… (2.46)

From thermodynamics; … (2.47)

G   RT ln K

So, the standard free energy of activation for the process O  J  OJ 

is;

G *   RT ln K *

And

  G *   K *  exp   RT 

... (2.47.a) ... (2.47.b)

And from Equation (2.46),  G *   RT   K   exp    Nh   RT 

... (2.48)

Again, from thermodynamic the relation of free energy to enthalpy and entropy is; … (2.49)

G  H  TS

So, Equation (2.48) may be written as;   H *   S *    RT       K   exp     RT R  Nh      

 H *   S *   RT      K  exp  exp   RT  R  Nh     

… (2.50)

A comparison of Equation (2.50) with Arrhenius Equation, the energy of activation Ea is related to the enthalpy of activation ΔH*. The pre S   RT   . The entropy of activation  exp   Nh   R 

exponential factor (A) is now 

will be negative for the reaction of the type O  J  OJ   Z , since OJ* is definitely more organized than O and J. For process of type OJ  OJ   O  J the

entropy of activation will generally be positive,

since activated complex will most likely have acquired some of the disorder which will eventually result in total break down into O and J. 24

Chapter Two


With more complex reactants, when the simple collision theory fails, Equation (2.50) is still satisfactory Peter and Julio, [2009]. According to historical application of the transition state theory on the corrosion of steel in acids, they stated that the rate-determining steps for hydrogen evolution reaction is the recombination of adsorbed hydrogen to form hydrogen molecules. In acid-free, the Transition State of the rate determining recombination step represents a more orderly arrangement relative to the initial state, and hence, negative value for the entropy of activation is obtained. It is widely recognized that the uniform CO2 corrosion rates are accelerated with the higher temperature where the governing kinetics are enhanced and the associated transport processes are accelerated. In addition, the tendency and rate of formation of the corrosion products, whose characteristics can control the corrosion reactions, are also temperature dependent Kermani and Morshed, [2003]. Temperature gives significantly effect to the corrosion rate, because an increase in temperature will cause a higher corrosion rate. However, temperature also accelerates the corrosion products which will be formed on the carbon steel surface and make a protective film. Based on previous experiment Agotnes et al., [1999], the protective properties of the film will improve when the temperature is increased. It is showed by temperature below 60°C, the film is easily removable, while a stable protective film is formed above temperature 60°C. According to Dugstad, [2004], the morphology of the surface films is temperature dependent; a) Below 40°C, surface films present an open porous structure and are formed mainly of Fe3C with some FeCO3 and alloying elements of the steel, b) at 60°C, the films present an inner porous part mainly of Fe3C with more FeCO3 accumulated in outer part. However, the formation of FeCO3 did not reduce corrosion rate 25

Chapter Two


significantly, c) at 80 °C, a dense protective FeCO3 film is formed close to the metal and it decrease the corrosion rate quickly. In that respect of corrosion kinetics and products, it was reported that the corrosion rate was proportional with temperature in a range from 25 to 60 °C, but it stabilized within the higher temperature range from 90 to 125°C attributing that to the physical factors induced by the corrosion products. Similarly, the temperature dependence of the corrosion rates was preserved in CO2-saturated media containing 1 mass% NaCl solution as the corrosion rates increased from 1 to 3 mm/yr, when the temperature increased from 20 to 80oC respectively Nesic, [1996]. In bicarbonate solutions, the electrochemical studies revealed that both anodic and cathodic reactions get accelerated with the higher temperature Li and Zuo, [2008]. Additionally, corrosion reactions at the low temperature 22 o

C were under charge transfer controlled while they were mass transfer

limit controlled at the higher temperatures, 40 and 80 oC. Temperature accelerates all processes involved in CO2 corrosion including transport of species, chemical reactions in the bulk of the solutions and electrochemical reactions at the metal surface. The growth of iron carbonate films is a very slow and a temperature dependent process. Increasing the temperature increases the precipitation rate of iron carbonate significantly. Depending on the solubility of protective films, temperature can either increase or decrease the corrosion rate. In the case of corrosion where protective films do not form (typically at low pH), corrosion rate increases with increase in temperature. However, at a higher pH increased temperature would accelerate the kinetics of precipitation and facilitate protective film formation, thus decreasing the corrosion rate Agotnes et al., [1999]. Temperature influences the conditions of protective iron carbonate layers. At temperature below 60oC, hydrogen evolution acts as a rate 26

Chapter Two


determining step and carbonate scale does not form well Fajardo et al., [2013]. Carbonate scale governs corrosion rate at the range temperature of 60 – 100 oC when protective films is formed. The greatest effect of temperature effect is attained by activation controlled processes George et al., [2004]. Temperature in general accelerates the process involved in corrosion, such as electrochemical, transport, crystallization etc. Nevertheless, the effect of temperature is also influenced by pH where the FeCO3 film may form and reduce corrosion rate. Nesic summarized that at low pH where the precipitation of iron carbonate or other protective scales does not occur, corrosion rate increases steadily with temperature. This effect can be related to the high solubility limit of FeCO3 and also the decrease in solution viscosity at higher temperature Nesic, [2007]. While at higher pH (more than 6), in the conditions where solubility of FeCO3 is low, an increase in temperature will enhance the kinetic of FeCO3 precipitation and protective film formation. Nesic et al. proposed a model that predicts the increase in precipitation rate constant as a function of temperature Nesic et al., [2003]. The precipitation rate constant is based on Arrhenius law, and defined in Equation (2.51) Sun and Nesic, [2008]: (

)

(

)

(

(

)

)

(

)

With A = 28.2 and B = 64.851 kJ/kmol. Moreover, some previous works overlooked that the increase in temperature is always accompanied by the increase in water vapour pressure, and lower CO2 partial pressure respectively Nesic et al., [2003]. Consequently, according to Henry’s law, the amount of dissolved CO2 in water is also decreased. Thus, the increase in temperature will increase

27

Chapter Two


the kinetics of precipitation, and also reduce the supersaturation of FeCO3 as a result of lesser amount of dissolved CO2 in the water.

2.9.2 Effect of pH: Normally, the higher pH will reduce corrosion rate. pH is the indication of the H+ concentration in the solutions, which is one of the main species involved in the cathodic reaction of CO2 process. The pH is influenced by changing the H+ ions concentration, temperature, pressure and ionic strength. The effects of the dissolved iron bicarbonate will also increase pH Fajardo et al., [2013].The increase of pH causes film become thicker, denser and more protective George et al., [2004]. It has been illustrated both experimentally and computationally that corrosion rate changes significantly with respect to pH. Higher pH leads to a decreased solubility of iron carbonate and thus results in an increased precipitation rate, faster formation of protective films and hence reduction of the corrosion rate. The control of pH is a common corrosion mitigation method and has been used with success in gas condensate pipelines Halvrsen and Anderson, [2003]. It consists of the increase of pH in the water phase by injecting sodium bicarbonate or sodium hydroxide. It facilitates the formation of a dense iron carbonate product on the metal surface which has good protective properties against corrosion. Film properties such as thickness, porosity, and adherence therefore become very important.

2.9.3 Effect of Speed of Rotation: Generally an increase in the speed of rotation of relative movement between a corrosive solution and a metallic surface tends to accelerate corrosion. This effect is due to higher rate at which the corrosive chemicals are brought to the corroding surface and to the higher rate at which corrosion products, which might otherwise accumulate and stifle 28

Chapter Two


corrosion, are carried away. The higher the speed of rotation, the thinner will be the film through which corroding substances must penetrate and through which soluble corrosion products must diffuse. Whenever corrosion resistance results from the accumulation of layers of insoluble corrosion products on the metallic surface, the effect of high velocity may be either to prevent their normal formation or to remove them after they are formed. The effects of velocity on corrosion rate are like the effect of oxidizer addition, complex and depend on the characteristics of the metal and environment to which it is exposed. Increased the velocity may increase or reduce attack, depend on its effect on the corrosion mechanism involved. Velocity may increase the diffusion or transfer of ions by reducing the thickness of the stagnant film at the surface. Velocity primarily affects on corrosion rate through its influence on diffusion phenomena. For corrosion processes which are controlled by activation polarization, velocity has no effect on the corrosion rate. If the corrosion process under cathodic diffusion control then the velocity increases the corrosion rate George and Nesic, [2007]. Velocity induced corrosion is due to combination effects of mechanical and electrochemical forces. The velocity results in thinner boundary layer which allows dissolved CO2 in water to corrode the surface more quickly. Higher velocity will also increase the wall shear stress that can cause localized corrosion and surface damages Silverman, [2005]. When the corrosion reaction is dominated by charge transfer, the increase in velocity has less effect on the corrosion rate. Increase in velocity decreases the precipitation rate and surface saturation of Fe +2 and CO3-2 because of near wall turbulence, which prevents Fe+2 ions from precipitating Nesic et al., [2003]. On the other hand, at low velocity, the

29

Chapter Two


rate of precipitation is higher than the corrosion rate thus enabling protective film formation. The effects of velocity on CO2 corrosion rate by increasing the transport of species between steel surface and bulk solution. This is particularly relevant in corrosion without film formation where higher velocity will increase corrosion rate. In the conditions with protective scales formed (higher pH), velocity may remove the film leading to an increase in corrosion rate Nesic, [2007]. However, as the main corrosion resistance in the presence of protective film is not only the species transfer but also the film layer itself; the effect of velocity is not as great as in the condition without film formation. According to study conducted by Zhang et. al., reduction of H2CO3 which is more pronounced at higher pH was found to be chemical reaction controlled and insensitive to velocity Zhang et al., [2006]. In addition, the limiting current for the cathodic reaction is not controlled entirely by diffusion, but also by slow hydration of CO2 that is not highly affected by the velocity. The corrosion rate would be directly proportional to the limiting diffusion current until the intersection of anodic and cathodic polarization curves occurs at a current less than the limiting diffusion current. At higher velocities, the corrosion rate would be relatively independent of velocity as shown in Figure 2.5.

30

Chapter Two


Fig. 2.5 Evans Diagram for Different Rotation Speed for Carbon Steel in Deaerated 1.0 M NaCl Solution at 50°C and pH 4 Andijani, [2002].

2.9.4 Effect of Acetic Acid (HAc) Concentration: Normally as the concentration of acid is increased, the corrosion rate is likewise increased. This is primarily due to the fact that the amounts of hydrogen ions, which are the active species, are increased, as acid concentration is increased. Hydrogen ion activity is commonly expressed, for convenience in term of pH (i.e., pH = - log [H+]). At low pH values hydrogen evolution usually predominate both in presence and absence of oxygen. The presence of HAc can change the mechanism of the anodic dissolution of iron through competitive adsorption of acetate ions, CH3COO- (or Ac-) and HCO3–. The carbonic acid (H2CO3) is not fully dissociated in the solution; it provides a reservoir of H+ ions which contribute the additional cathodic reactions. The increase of corrosion rate also attribute to the forming of iron acetate George and Nesic, [2007] and Tran et al., [2013].

2.10 Corrosion Product Film Formation: CO2/HAc corrosion on the metal surface is strongly dependent on the type of corrosion product film formed on the surface of the metal during the corrosion process. The precipitation rate or the formation of these films depends on various environmental factors and greatly on the 31

Chapter Two


concentration of species. The stability, protectiveness, and adherence of these films determine the nature and the rate of corrosion. Depending on the composition, the corrosion films can be of different forms.

2.10.1 Iron Carbide (Fe3C): Iron carbide is an undispersed component of mild steel, which is left behind after the corrosion of iron from the steel structure. Iron carbide films are conductive electrically, very porous and non-protective. These films can significantly alter the corrosion process by either decreasing the corrosion rate by acting as a diffusion barrier, or increasing the corrosion Gulbrandsen et al., [1998] by increasing the active specimen surface area by forming a conductive bridge between the counter and working electrodes. Also, this kind of film formation could result in galvanic coupling of the film to the metal or acidification of the solution inside the corrosion product film which is very dangerous and by far the strongest reason that could be given for the occurrence of localized corrosion.

2.10.2 Iron (II) Carbonate (FeCO3): Iron (II) carbonate (FeCO3) is an insoluble corrosion product which forms a film that potentially can be act as protective layer on the corroding surface. According to the previous experiments, Iron (II) carbonate (FeCO3) is important in the formation of protective layers Dugstad, [2004]. The equilibrium that describes the formation of iron (II) carbonate is:



(

( )

)

The precipitation rate determines the scale growth and its protectiveness of FeCO3 because when FeCO3 precipitates at the steel surface, the corrosion process can be slow down by Nesic, [2007]:  Presenting a diffusion barrier for the species involved in corrosion process. 32

Chapter Two


 Covering (inhibiting) a portion of the steel surface. FeCO3 can precipitate not only on the steel but also directly on the Fe 3C as a result of the ambient concentration in Fe2+ and the additional HCO3anions produced on Fe3C by the cathodic reduction of CO2. The protective and non protective layers are depends on the presence and absence of Fe3C in contact with steel; if Fe3C is presence and in contact with steel, then the layer is non protective. On the other hand, if Fe3C is absence, then the layer is protective. The non protective and protective layers are shown in Figure 2.6.

Fig. 2.6 Morphologies Observed for Protective and Non Protective Corrosion Layers Crolet et al., [1998].

2.11 Electrochemical Measurements: 2.11.1 Polarization: In electrochemical corrosion testing two direct approaches are apparent: control the current (i.e., corrosion rate) and measure the resulting potential, or control the potential (i.e., oxidizing power) and measure the resulting current. In each case the potential of an electrode in a conducting media is changed by the flow of current in the electrolytic 33

Chapter Two


cell. This change in potential from a reversible or steady state value as a result of current flow is known as polarization. The schematic curves that are shown in Figure 2.7 illustrate some of the terminology, are plotted in accordance with ASTM Recommended Practice. The anodic polarization curve is represented as a metal dissolution reaction ( M  M 2  2e ) which is predominant at potentials more positive (noble) than the specimen open circuit corrosion potential (Ecorr), and the cathodic polarization curve is represented as a reduction reaction ( 2H   2e  H 2 ) or O2  2H 2O  4e  4OH  at potentials more negative (active) than Ecorr.

Fig. 2.7 Polarization Diagram illustrates some of the Terminology Uhlig and Winston, [2008]. There are three fundamental types of anodic and cathodic polarization,

namely,

activation,

concentration,

and

resistance

polarization. Under activation polarization the reaction sequence at the metal-electrolyte interface controls the electrochemical process. One example is the corrosion that occurs in media containing a high concentration of active species, such as relatively concentrated acids. 34

Chapter Two


Under activation control, anodic and cathodic data for potential versus the logarithm of the applied current density give linear behavior when the amount of polarization, which is called overvoltage, is more than about 50 mV from the open circuit corrosion potential. For the anodic curve the dashed line in Figure 2.7 represents the corrosion rate variation with potential. The deviation of the observed curve (solid line) from the theoretical at potentials within 50 mV of the corrosion potential is due to the fact that the applied current density (iapplied) is the absolute difference between the total oxidation current density (iox) and the total reduction current density (ired). Near intersections of the oxidation and reduction curves this subtraction is significant, but at about 50 mV beyond the intersection potential, for cathodic curves iapplied = ired for anodic curves iapplied = iox and iox= icorr. at each potential 50 mV more noble than Ecorr as long as corrosion is the predominant oxidation reaction. When the reaction rates are controlled by the diffusion of species in the bulk electrolyte to the metal-electrolyte interface, concentration polarization is observed. This behavior usually occurs when the concentration of reducible species in the environment is small, for example, corrosion in dilute acids and in aerated salt solutions. In such cases there is a precipitous change in potential at the limiting diffusion current density as illustrated in Figure 2.7. Resistance polarization can result from an IR potential drop during electrochemical measurements in low—conductivity solutions. This phenomenon has not been a serious factor in corrosion Studies. Additional theoretical background on polarization, the mixed potential theory for electrochemical corrosion, and deviations of observed plots from linearity near the open circuit corrosion potential can be found in modern textbooks. Since the electrical data, either current or potential, are amenable to precise measurement, controlled polarization has become an important 35

Chapter Two


experimental technique in corrosion testing. By measuring the impressed current density as a function of potential, the corrosion resistance of a specimen can be estimated over a wide range of oxidizing conditions, On the other hand, one may select a specific potential and polarize specimens to that potential for a fixed time and then determine the extent of corrosion attack Uhlig and Winston, [2008].

(i) Activation Polarization (ηA): This is the polarization caused by slow electrode reaction, or started in another way; the reaction at the electrode requires activation energy in order to go. The most important example is that hydrogen 1 2

reduction at a cathodic, H   H 2  e , the corresponding polarization term being called hydrogen overpotential. The relation between reaction rate and overvoltage for activation polarization is Buchanan and Stansbury, [2000]:  A  β log

i io

… (2.53)

Where: ηA is overpotential (overvoltage) due to activation β is Tafel constant i is the rate of oxidation or reduction in terms of current density. Equation (2.53) is called the Tafel equation and β is frequently termed "β slope" or Tafel constant. Equation (2.53) is graphically illustrated in Figure 2.8. The value of β for electrochemical reactions is 0.05 to 0.15 volt. The oxidation and reduction reactions corresponding to a hydrogen electrode are plotted with a β value of 0.1 volt. Note that the reaction rate changes by one order of magnitude for each 100 mV or 0.1 volt change in overvoltage. It can be seen that at all potentials more noble than the reversible potential, a net oxidation process occurs and that at all 36

Chapter Two


potentials more active or more negative than the reversible potential, a net reduction occurs. At reversible potential, or at zero over voltage, there is no net rate of oxidation or reduction since both rates are equal at this intersection point. From the Equation (2.53), the larger the value of io and the smaller the value of β, the smaller is the corresponding overpotential. Overvoltage, V 0.2 Noble

H 2  2H   2e 

ioH2/H+

0.1 0

2H   2e  H 2

Active -0.1 -0.2

Current density 0.1

1

10

100

Fig. 2.8 Activation Polarization Curve of a Hydrogen Electrode Uhlig, [2013].

(ii) Concentration Polarization (ηC): The rate of an electrode reduction also depends on mass transfer i.e., the rate at which the reactant is transported to the electrode and the rate at which the product is transported away from the electrode. Transport through the solution to and from the metal surface occurs by the diffusion, ion migration (transport of electrical charge through the solution), and convection. Of this diffusion through the thin layer of the solution adjacent to the metal surface, the diffusion layer δd, is usually of greatest significance. However, this is not always the case in the practical systems, particularly where dissolved oxygen is the cathodic reactant and in certain circumstances the rate of diffusion through the bulk solution to 37

Chapter Two


the metal-solution interface may be rate determining. The limiting current density (the maximum possible rate per unit area under the conditions prevailing) for cathodic process is given by McCaffrty, [2010]: … (2.54) ilim 

DzFCb

d

The Equation for concentration polarization is given by Philip et al., [2002]: … (2.55) C 

RT  i  RT C s  ln 1  ln zF  ilim  zF Cb

… (2.56) C 

, at 25 oC

0.059  i   log1  z i lim  

Where: ηC is overpotential (polarization) due to concentration, (V) R is universal gas constant, 8.314 J/gmole. K T is temperature in Kelvin (K) F is Faraday's constant, 96494 C/eq. z is valency. Cb is concentration of reactant in the bulk, (mol/lit) Cs is concentration of reactant at the surface of electrode, (mol/lit) And it is evident that the smaller the ilim the greater the magnitude of the overpotential due to concentration (transport). Unlike activation polarization, concentration polarization is not controlled by the kinetics of charge transfer and the magnitude of ηC will be the same for any cation (providing z, D and Cb are the same) and the metal surface. Thus the ratecontrolling parameter in the concentration polarization is ilim , and it will be seen that any factor in Equation (2.54) that causes ilim to increase will result an increase in corrosion rate, providing that latter is solely determined by the kinetics of the cathodic process Sharma, [2012].

38

Chapter Two


Generally, concentration polarization changes are not problem with the anodic reaction. However, concentration polarization is often factor in determining the rate of cathodic reaction. Concentration polarization occurs when one of the reactants consumed at an electrode surface is faster than it supplied from the bulk of the solution. The rate of the reaction is limited by diffusion from the solution to the electrode surface Robert et al., [2003]. A graphical representation of the Equation (2.55) is shown in Figure 2.9. Concentration polarization does not become apparent until the net reduction current density approaches the limiting diffusion current density. Examination of Equation (2.55) indicates that when the net reduction current is equal to the limiting diffusion current, overvoltage is equal to infinity. + ilim ηC

0

-

log i

Fig. 2.9 Concentration Polarization Curve (Reduction Process) McCaffrty, [2010].

(iii) Resistance Polarization (ηR): The resistance polarization is the ohmic potential drop through a portion of the electrolyte surrounding the electrode, through a metal reaction product film on the surface or both. An ohmic potential drop always occurs between the working electrode and the capillary tip of 39

Chapter Two


reference electrode. The ohmic (i.e., solution) IR drop is given by Robert et al., [2003]: … (2.57)

IR soln.  i ρ l  i 1 k

Where: ρ = The specific resistance (i.e. resistivity) (Ω.cm) k = =The conductivity (Ω-1.cm-1) or (S.cm-1); S = Siemens l = The solution gap between the capillary tip and the working electrode (cm). i = The current density (A/cm2) Resistance polarization is important only at high current densities or in high resistance electrolyte solution Einar, [2004]. All of these three types of polarization will be present to a greater or less extent in most corrosion reactions.



Total

 A  C  R

… (2.58)

But if one is more influential than the others, then it will control the reaction rate.

(iv) Combined Polarization (ηT): Both activation and concentration polarization usually occur at an electrode. At low reaction rates, activation polarization usually controls, while at higher reaction rates concentration polarization becomes controlling. The total polarization of an electrode is the sum of the contributions of activation polarization and concentration polarization: T   A  C

… (2.59)

Where ηT is total overvoltage. During anodic dissolution, concentration polarization is not a factor as mentioned above, and the Equation for the kinetics of anodic dissolution is given by:

40

Chapter Two


i io

… (2.60)

 diss  β log

During reduction process such as hydrogen evolution or oxygen reduction, concentration polarization becomes important as the reduction rate approaches the limiting diffusion current density. The overall reaction for a reduction process in given by combining Equations (2.53) and (2.55) with appropriate signs:  red  β log

 i RT i  2.3 log1  io zF  ilim

  

… (2.61)

Equation (2.61) is graphically illustrated in Figure 2.10. The importance of Equations (2.60) and (2.61) cannot be overemphasized since they are the basic Equations of all electrochemical reactions. Equation (2.61) apply to any reduction reaction and Equation (2.60) applies to almost all anodic dissolution reactions. Using only three basic parameters, namely, β, io and ilim , the kinetics of virtually every corrosion reaction can be precisely described. Equations (2.60 and 2.61) represent an outstanding simplification of the complex phenomena observed during corrosion reactions Einar, [2004]. + io ηT

0

Activation polarization Concentration ilim polarization

-

log i

Fig. 2.10 Combined Polarization Curve-Activation and Concentration Polarization McCaffrty, [2010].

41

Chapter Two


2.11.2 Mass Transfer: When an electrode is polarized, the surface concentration of species that is either being oxidized or reduced fall to zero. Additional material will then diffuse to the electrode surface towards this region of lower concentration. If the electrolysis experiment is carried out in an agitated solution, i.e. the solution is in motion with respect to the electrode or vice versa, then the resulting concentration-distance profile at the electrode surface can be represented as shown in Figure 2.11. In the Figure 2.11 the electrode is represented by bold bar in the left side of Figure. The x-axis is a scalar axis and represents distance away from the electrode. The point of origin (x = 0) represents the surface of the electrode. The y-axis represents concentration. The maximum concentration is represented by Cb which is its concentration in the bulk solution. Electrode δd

Bulk solution C = Cb

Concentration

x=0

x = δd

Distance from electrode

Fig. 2.11 Variation of Anion Concentration with Distance from Electrode Roberge, [2004].

42

Chapter Two


In Figure 2.11, there are two regions of concentration. Because the solution is well mixed, in the bulk region the concentration is constant with respect to distance. This is represented by the horizontal line where C = Cb (this is known as the convective region). There is then a region where the concentration drops, falling to zero at the electrode surface. The diffusion layer associated with this drop has thickness δd. The exact thickness of the diffusion layer depends upon the nature of the solution into which it extends. For stirred aqueous solution the thickness of the diffusion layer can be between 0.01 to 0.001 mm Roberge, [2004]. For an unstirred solution it is about 0.5 mm thick. In an important assumption of this mode is that when material reaches the surface of the electrode it is instantaneously oxidized or reduced thereby maintaining a zero concentration at the electrode surface. In practice, this is easy to achieve by selecting a suitable polarizing voltage. It is clear that the concentration gradient in the diffusion layer, we note that it is both linear and constant. It can be representing gradient mathematically as: dC C b  C s   dx d

… (2.62)

Fick's law of diffusion state that Perez, [2004]: J  D

dC dx

… (2.63)

Where J is the flux of substance and D is a diffusion coefficient. It can be seen from faraday's law of electrolyte that: … (2.64)

i  zFJ

Substituting Equation (2.63) into Equation (2.64) gives: i   zFD

dC dx

… (2.65)

Substituting Equation (2.62) into Equation (2.65) gives:

43

Chapter Two

i   zFD


Cb  C s 

… (2.66)

d

The negative sign is conventional, it tell us the current is being carried away from the cathode (as hydroxyl ion). The magnitude of current is interesting only and since, Cs can never be negative, and the magnitude of the current is greatest when Cs=0, again the maximum or limiting current density: i   zFD

Cb

… (2.67)

d

Notice that when Cs > Cb, the implication is that the concentration of species in the region of electrode is increasing. The resulting sign from negative to positive tells us the current is reversed Cottis, [2002].

2.11.3 Corrosion Rate Measurements: Besides the weight loss technique there are other techniques used in corrosion rate measurements.

(i) Linear Polarization Technique: The linear polarization technique in principle is a convenient and rapid way for determining corrosion rates. Although this technique was first developed some sixty years ago by Wagner and Traud, [1938] who stated that the corrosion process with two coupled electrochemical reactions under activation control can be represented by:   i  iCorr. exp   ba 

     exp     bc

  

… (2.68)

But it lay dormant until the early nineteen-fifty when it was used by Stern and Geary, [1957] in various studies of metal dissolution in acids. Briefly their method rests on the mathematical consequence of expanding two exponential functions around the rest potential Equation, (2.68), for relatively small increment of the exponent (i.e., for potential within about 10 mV of the corrosion potential). 44

Chapter Two


It is observed that the applied current density is a linear function of the electrode potential. The slope of this linear polarization is related to the kinetic parameters of the system as follows ba bc E   iapp. 2.303 icorr. b a  b c 

… (2.69)

Where; b a & b c are the anodic and cathodic Tafel slopes respectively. Robert et al., [2003] pointed out that for the general case it was unreasonable to treat polarization curve in the region near the corrosion potential as a straight line. They demonstrated that the inflection point of the polarization curve did not coincide with the point ΔE = 0 except in the case where Tafel slopes of anodic and cathodic reactions of the corrosion process where equal (i.e., b a  b c ). In addition they noticed that, in general, it was unsuitable to take (10 mV) as the confines of linearization, because in most cases polarization curves would obviously deviate from the straight line before

E reached (10 mV). Thus Equation (2.69) will bring a high percent of error because of no linearity near the corrosion potential.

(ii) Tafel Extrapolation Technique: The Tafel extrapolation method uses data obtained from either cathodic or anodic polarization measurements. If the potential of electrode (with respect to the reference electrode) is plotted against the logarithm of applied current then a figure similar to that shown in Figure 2.12 is produced.

45

Chapter Two


Fig. 2.12 Tafel Extrapolation Technique on Cathodic Region Uhlig and Winston, [2008]. The applied current polarization curve indicated by points and a solid line. At low currents the curve is non-linear; but at higher current it becomes linear on a semi-logarithmic plot. Applied current is related to cathodic and anodic current as:

i app.  i red.  i oxd .

… (2.70)

At high-applied currents the value of the anodic current is small compared with cathodic current so at high-applied current it begins to approach total actual cathodic current. In actual practice, an applied polarization curve becomes linear on a semi-logarithmic plot at approximately (50 mV), more active than the corrosion potential. This region of linearity is referred to as the Tafel region. To determine the corrosion rate from such polarization measurements, the Tafel region is extrapolated to the corrosion potential. At the corrosion potential, the rate of anodic and cathodic reactions is equal. This point corresponds to the corrosion rate of the system expressed in terms of current density. This method can only be applied to system containing one reduction process, since the Tafel region is usually 46

Chapter Two


distorted if more than one reduction process occurs. This technique has been used by many investigators to measure the corrosion rate in CO2 saturated, NaCl solutions of protected and unprotected mild steel.

(iii) Effect of Concentration Polarization on the Determination of Corrosion Rates from Polarization Measurements: The generally used methods for calculation of corrosion rates from polarization measurements are based on the assumption (among others) that the process is controlled by the rates of anodic and cathodic partial reaction and the supply of the reactants is unlimited. These methods use a polarization equation that is derived for the complete absence of masstransport effect; occasionally, an equation that considers the cathodic partial reaction to be under complete mass-transport control is also used. The mixed-control case, when the rate of the process is jointly controlled by the surface reaction and the transport of material to and from the surface, has been much less investigated, even though this case may quite frequently reflect practical situations. Nagy and Thomas, [1986] derived the following Equation which can be used when the cathodic reaction is under mass-transport control (icorr= ilim ), mixed control (icorr< ilim ) , or charge-transfer control ( ilim = ∞)   2.303E   2.303E     exp    i  icorr exp  b b a c     

...(2.71)

Where



1

(2.72)

 2.303E  i i  1  corr  corr exp   ilim ilim b c  

This Equation can be used as a basis for curve-fitting evaluation and for error analysis of the corrosion rate from polarization measurements.

47

Chapter Two


McLaughlin, [1981] has derived the following polarization Equation which is corrected for mass-transport effect and contains only one additional parameter, the limiting current density. (

)

(

where is applied current density,

)

is the corrosion current,

(

)

and

are the Tafel slopes for the anodic and cathodic process respectively, , where

is the corrosion potential and

electrode potential corresponding to the applied current,

is the and

are diffusion currents for the anodic and cathodic processes respectively. The total derivation of Equation (2.73) was given in the literature McLaughlin, [1981]. He has proposed this Equation as a basis of a curve-fitting evaluation of the corrosion rate from polarization measurements.

2.12 Research Methodology of Building an Experimental Design and Optimization for CO2 Corrosion: Relationship between corrosion rate and factors involved in CO2 environment has been published extensively by George and Nesic, [2004] and George and Nesic, [2007]. The reports stated that the corrosion rate has polynomial model regression. The polynomial model regressions usually occur when protective film is formed Asmara and Mokhtar, [2008]. Therefore, a suitable design to estimate a simple curvature model regression in wide range of temperature is on an assumption that corrosion rate (response) will behave a second-degree polynomial model. So, the first order model is definitely not suitable to fit the setting parameter. Then with the aim to optimally research, design experiment response surface was applied. The minimum number of levels required for each factor to quantify that behavior is two. Further, by adding center 48

Chapter Two


points with some repetition variants would satisfy the requirement for pure error analysis Devore, [2005]. A problem which often occurs in the design of an experiment in physical or industrial research is that of determining suitable tolerances for the components of a certain assembly; more generally of ascertaining the effect of quantitative or qualitative alterations in the various components upon some measured characteristic of the complete assembly. It is sometimes possible to calculate what this effect should be; but it is to the more general case when this is not. In such case it might appear to be best to vary the components independently and study separately the effect of each in turn. Such a procedure, however, is wasteful either of labor or accuracy, while to carry out a complete factorial experiment (i.e. to make up assemblies of all possible combinations of the

components) would require Lk assemblies, where

(L) is the number of values (assumed constant) at which each component can appear. For L equal to 2 this number is large for moderate ( ) and quite impracticable for ( ) greater than say, 10. For larger L the situation is even worse. What is required is a selection of N assemblies from the complete factorial design which will enable the component effect to be estimated with the same accuracy as if attention had been concentrated on varying a single component throughout the (N) assemblies Jeff Wu and Michael, [2009]. In the chemical industry experimental designs are particularly applied to the study of process variables and how they affect the product. Perry and Green, [2000] Experimental variables are usually called factors. The particular value of the variable is called the level of the factor. The combination of 49

Chapter Two


factors used in a particular experiment run is called a treatment. The result of the run is designated as the effect. Three basic types of statistically designated experiments are most often used in chemical industry Robert and James, [2003]: 1. Factorial design. 2. Fractional factorial design. 3. Box-Wilson design. Factorial design is used as the first step to determine the influential factors and use regression analysis on further experimental runs to develop a quantitative relationship among the important factors; it can be applied to any number of factors and levels. Fractional factorial utilizes some integer fraction (a multiple of the number of levels) of the corresponding total factorial experiments. Box-Wilson design is a series of tests for characterizing a physical mechanism. These series of experiments have been developed which efficiently serve as a basic for deriving the mathematical model of process. Their usefulness is enhanced in the study of industrial application because most physical situations can usually be approximated by a quadratic function over a reasonable range of the factors. A preliminary step is to setup the relationships between coded level and corresponding real variables. These relationships are as follows Tarantino, [2010] and Forsal et al., [2010]: ( Where:

)

(

= value of the variable factor in i-coded;

value of factor i in natural variable; and

)

= corresponding

= central value in the field of

variation: ( 50

)

(

)

Chapter Two


where k is the number of the input factors. The number of experiments required, N, according to factorial method is: (

)

Factorial points A

factorial design seemed to be the natural choice for this

experimental situation. It was chosen for the following reasons Khattree and Rao, [2003]. 1. The factorial design is an efficient method of experimentation. It provides information on the effect of several variables almost as quickly as a comparable amount of information can be generated on the effect of one variable alone. 2. It provides a measure of interaction between controlled variables if such interaction exists in fact (A system containing interaction exhibits curvature in its response to changes in levels of the independent variables). 3. An additional check on the existence of curvature can be provided by adding a center point to the design. The difference between the response at the center point and the mean for all of the peripheral points is a measure of "lack of fit ". Lack of fit refers to the inadequacy of a linear model to represent the data. 4. The factorial design can be augmented at a later data to provide an estimate of curvature should it be found to exist. This allows the experiment to proceed sequentially first with a relatively simple and efficient experimental design, and later- but only if necessary –with one of greater sophistication. 5. The experiment may be kept within a practical size limit by running the treatment combination in balanced blocks, Jeff Wu and Michael, [2009].

2.13 Regression Analysis: 51

Chapter Two


Various theoretical models that have been proposed are not accurate enough and can be applied only to a limited range of processes and corrosion conditions. For these reasons, most researchers mainly use the empirical research. The regression analysis technique, based on the experimental data, is a powerful tool for modelling and analyzing real processes, whose nature and behavior cannot be explained using a theoretical approach. Many researchers use this method successfully in various fields. Therefore, with the efficient regression analysis researchers cannot only rely on their perspicacity and intuition, but must have a relevant knowledge of the researched phenomenon and the experimental techniques. Many experiments involve studying the effects of more factors. In these cases, generally, the design of experiment (DoE) is the most efficient type of experiment, especially in relation to the traditional onefactor-at-a-time experiment. The selection of a proper experimental design is essential for reducing the experimental cost and time. The success of a regression analysis depends largely on the choice of appropriate mathematical models. Many studies have shown that the choice of mathematical models in the form of polynomials provides the most appropriate and effective approximation of the experimental data. These are the following mathematical models: 1) Linear mathematical model ∑

̂

(

)

2) Quasi-linear mathematical model ∑

̂

∑

∑

(

)

(

)

3) Non-linear (quadratic) mathematical model ̂

∑

∑

∑ 52

∑

Chapter Two


where ̂ is the estimated response, ye is the measured response, εi is the independent random variable (experimental error), b0 is the free term (parameter) of the mathematical model, bi are the linear terms, bii are the quadratic terms, bij are the interaction terms, and k is the number of the independent variables (factors). The parameters of the mathematical model can only be statistically estimated on the bases of the experimental results. The relationship between dependent variable (response) and independent variables (factors) can also be expressed in the form of the multiple power function: ∏

(

where Y is the estimated natural response,

and

)

are constants to be

estimated. Applying the logarithmic transformation, the non-linear Equation (2.78) can be converted into the following linear Equation: ∑

(

)

When the variables in logarithmic scale in Equation (2.79) are replaced with the new variables, ̂

,

(

), then it can be

rewritten in a linear form, defined by Equation (2.75). If the multiple power function includes first-order factor interactions, then Equation (2.79) represents the quasi-linear mathematical model, defined by Equation (2.76). But, in this particular case, it is not necessary Lazarevic et al., [2010]. The relationship between dependent variable (response) and independent variables (factors) can also be expressed in the form of the exponential and multiple power function: (

( ⁄

))

(

53

( ⁄

)) ∏

(

)

Chapter Two


where Y is the estimated natural response,

,

and

are

constants to be estimated. Applying the logarithmic transformation, the non-linear Equation (2.80) can be converted into the following linear Equation: ( ⁄ ( ⁄

)

) ∑

(

)

In general, the mathematical models may also include higher-order factor interactions. Since the impact of higher-order factor interactions is usually negligible, these terms of the mathematical model may be omitted. On the other hand, in many cases, adding the high-order polynomial terms does not really improve the fit, but increases the complexity of the mathematical model. Thus, it is useful to try fitting using a lowest-order polynomial that adequately describes the system/process. The statistical method often used to estimate the unknown parameters in a mathematical model is the method of least squares. The number of factor levels within the selected range is theoretically arbitrary, whereas practice confirms that it is sufficient to choose: two levels for (quasi) linear mathematical model and non-linear (quadratic) mathematical model. Since input factors may be various physical values (temperature, pressure, volume, velocity, etc.) it is useful to perform their coding. The most useful application for DoE is to optimize a process/system. The process optimization is assured by minimizing or maximizing an objective function regarding the given (in) equality constraints. The second-order mathematical model may be written in matrix notation as following Montgomery, [2005]: ̂

̂

‫׳‬

(

‫׳‬ 54

)

Chapter Two


where x is a (k x 1) vector of the independent variables, b is a (k x 1) vector of the first – order regression coefficients and B is a (k x k) symmetric matrix whose main diagonal elements are the pure quadratic coefficients, while off-diagonal elements are one-half mixed quadratic coefficients. The stationary (optimal) point is obtained from the following relation: (

)

Thereby it implies that the optimum conditions of objective function are met. Furthermore, by substituting Equation (2.83) into Equation (2.82) the predicted response at the optimal point can be found as: ̂

̂

(

)

For the characterization of the corrosion rate response (output) it is necessary to translate the selected mathematical model into a canonical form. Canonical transformation transfers the origin in the stationary point and rotates the coordinate axis to match with the main axes of the fitted corrosion rate response. Canonical form of quadratic mathematical model can be expressed as follows: ̂

∑

̂

(

)

where {xi} are the canonical independent variables (factors) and {θ i} are their eigenvalues (canonical coefficient). Canonical coefficients are the roots of the characteristic Equation: |

|

(

)

(

)

Checking the correctness of calculation is done according to: ∑

∑

55

Chapter Two


Canonical Equations contain no linear effect or interactions, which makes them more suitable for the analysis of the surface response. The geometric form of the response surfaces is determined by the stationary point, algebraic signs, and magnitudes of their own values. If the eigen values are all negative, the response surface has a maximum; if they are all positive, the response surface has a minimum; if they have mixed signs, the response surface has a saddle point Jeff Wu and Michael, [2009].

2.14 Modeling and Theoretical Analysis of CO2 Corrosion System: In present work Equation (2.71) were used to determine the mechanism of corrosion of API X65 mild steel in CO2 saturated, 3.5 wt % NaCl solution in presence and absence of acetic acid. The approaches were suggested by Korobove and Medvedeva, [2000] will apply to limit whither the corrosion process be under charge-transfer, mass-transport or mixed control. The general methods used for calculation of corrosion rates from polarization measurements are based on the assumption that the process is controlled by the rate of anodic and cathodic partial reactions and supply of reactants is unlimited, and the mixed-control case, when the rate of the process is jointly controlled by the surface reaction and the transport of material to and from the surface has been much less investigated, even thought this case may quite frequently reflect practical situations. So, it is very important to study the effect of neglecting the mass transfer on corrosion parameters. The basic electrode kinetic Equation, which is derived in many references for general redox reaction, O + ne = R, is 56

Chapter Two


usually written as Bockris and Reddy, [1977] and Kreyszig, [2011]:

 i  io exp  s

 α  F η ct   α  F η ct     exp   R T R T    

… (2.88)

Where o

α n

α n

  C 

i  nFk C s o

s o

s R

… (2.89)

α  α  n

… (2.90)

Equation (2.88) is useful for conditions when mass-transport effect is negligible, that is, the surface concentrations are equal to the bulk concentrations and the measured over potential is equal to the chargetransfer over potential. If mass-transport effect is not negligible, Equation (2.88) has to be modified. Two modifications are needed: (1) the measured over potential must be corrected for the mass-transport over potential, and, (2) the exchange current density has to be expressed as a function of bulk concentrations. After taking into account this two-assumption, Equation (2.88) can be re-derived to yield Kreyszig, [2011],   2.303E   2.303E     exp    i  icorr exp  bc      ba

… (2.71)

Where



1

… (2.72)

 2.303E  i i  1  corr  corr exp   ilim ilim b c  

Where λ is the mass transfer correction factor, which can be written in the form,



1  2.303E   1     exp   b c  

… (2.91)

57

Chapter Two




icorr , and ilim is the limiting diffusion current. ilim

Equation (2.71) is completely general in that it describes the polarization curve for the case when the cathodic reaction is under masstransport control (i lim = icorr), mixed control (i lim > icorr), or charge-transfer control (i lim = ∞). If the exponential terms of Equation (2.71) are expanding in series, then the Equation of polarization curve become as an Equation of polynomial

Korobove

and

Medvedeva,

i   Co  C1  C2 2  C3 3  ...  Cn n

[2000]

as:

… (2.92)

Where Co, C1, C2, …Cn are constants. Equation (2.92) is a non-linear Equation of   i  relationship. The coefficients of   can be obtained by a non-linear estimation method, using MATLAB v.12 (Matlab Professional) Software Program , from a set of experimental points of i as a function of η for API X65 mild steel corrosion in CO2 saturated, 3.5 wt% NaCl solution in presence and absence of acetic acid (i.e., without and with the protective film formation). Over potential interval –100 mV to 100 mV from Ecorr. Equation (2.92) can be written in the form of the Maclaurin Formula as Kreyszig, [2011];  i  1   2i     2 i    2!      0

 1   ni 2    ...   n n!     0

   n   0

… (2.93)

By comparison of Equations (2.92) with (2.93), the polynomial coefficient can be written as; … (2.94)

Co  0  1 C1  2.303icorr   b a

  1      bc

  1     

… (2.95)

58

Chapter Two

  1 2C 2  5.303icorr  b   a

CO2 Corrosion of Mild Steel 2

  1      bc

  1 6C3  12.214icorr  b   a

  1      bc

  1 24C 4  28.13icorr  b   a

  1      bc

3

4

2

  1  3  2 2 



3





  1  7   12 2  6 3 



4

… (2.96)

  



… (2.97)

 

  1  15  50 2  60 3  24 4 





  

… (2.98)

The values of C1 to C4 of Equation (2.92) are listed in Appendix E, and the derivations of Equations (2.94 through 2.98) are listed in Appendix F. The sets of above Equations can be solved with the help of MATLAB Program-Substitution Method v.12 (Matlab Professional). The substitution procedure was listed in Appendix G. Corrosion process kinetic parameters determined by this model are listed in Tables (4.17 through 4.20) for the corrosion of API X65 mild steel in CO2 saturated, 3.5 wt% NaCl solution in presence and absence of acetic acid (i.e., absence and presence of the protective film formation).  icorr   or   can be used to determine the  ilim 

The dimensionless factor 

corrosion mechanism. The values of β have two limiting cases;   0 the corrosion kinetic is subjected by slow charge transportation; and at   1 the limiting stage of total process is the depolarizer transport to electrode surface. Korobove and Medvedeva, [2000] suggested that in the range of β < 0.05 and β > 0.95, it is possible to use the simplified form of Equations (2.71) and (2.94 through 2.98) (for activation and diffusion mechanism) without precision of calculations evaluation.

59

Chapter Three Experimental Work

Chapter Three

Experimental Work

T

he corrosion behavior of API X65 mild steel alloys, used widely in oil and gas production and transport industries, was studied

using weight loss and polarization technique in presence and absence of acetic acid in 3.5 wt. % NaCl of CO2–saturated solution by bubbling CO2 gas ( > 99.99%) at 1 bar for 1.5 hours, at different temperatures (40, 50 and 60 ºC), different pH’s (3, 4 and 5), different HAc concentrations (1000, 2000 and 3000 ppm) and different speeds of rotation (1000, 1250 and 1500 rpm). In absence of acetic acid at different temperatures (65, 70 and 75 ºC), different pH’s (7.5, 8 and 8.5) and different speeds of rotation (1000, 1250 and 1500 rpm).

3.1 Materials: 3.1.1 Corrosive Solution: The corrosive solution used in this work was 3.5 wt. % NaCl saturated with pure carbon dioxide CO2 gas during 1.5 h prior the immersion of the metal samples. The specification of the NaCl salt (Laboratory NaCl, Gainland Chemical Company GCC, England) contains the following materials:Chemical Formula

Concentration

Sodium Chloride

NaCl

99.9 %

Calcium

Ca

0.02 %

Barium

Ba

0.01 %

Iron

Fe

0.01 %

Silver

Ag

0.004 %

Material

Photograph

MW of NaCl Salt

58.44 g/mol

60

Chapter Three

Experimental Work

The chemical components of the simulated brine solution that used in weight loss technique compared with 3.5 wt. % NaCl solution are given in Table 3.1. Table 3.1 Chemical Components of the Simulated Brine Solution. Chemical Formula

wt %

Grade

Supplier

Sodium Chloride

NaCl

80.267

Laboratory

England GCC

Calcium Chloride

CaCl2

15.204

Laboratory

England Fluka

Magnesium Chloride

MgCl2

2.604

Laboratory

England BDH

Potassium Chloride

KCl

1.926

Laboratory

England Fluka

Material

Photograph

3.1.2 Carbon Dioxide CO2 Gas: High grade quality pure carbon dioxide CO2 gas (> 99.99%) was supplied by AL-Mansour factory in Baghdad. It is used in the beginning of experiment in order to de-oxygenate, saturation of the corrosive solution and the CO2 injection was maintained during the test with the aim of avoiding any oxygen contamination.

3.1.3 Chemicals: The Table 3.2 below lists the compounds and chemicals used in the investigation.

61

Chapter Three

Experimental Work

Table 3.2 Chemicals Used in this Investigation. Material

Photograph

Chemical Formula

Purity wt%

Grade

Company

Acetic Acid

CH3COOH

99.5

Laboratory

Germany RiedelDe Haen AG

Hydrochloric Acid

HCl

98.9

Laboratory

England Fluka

Ethanol

C2H5OH

99.99

Laboratory

England Fluka

Sodium Bicarbonate

NaHCO3

97.9

Laboratory

England BDH

SiO2

99

Laboratory

India Thomass Baker

FeCl2

98

Laboratory

England BDH

Silica Gel

+ H2O

Ferrous Chloride

3.1.4 Working Electrodes: The material of electrodes used in this investigation was a mild steel pipe (API X65), supplied by Oil Pipelines Company (OPC) in AlDaura Refinary in Baghdad. The specimens were of dimensions 2.03cm outside diameter, 2.08 cm long and 0.13 cm thickness, so an overall surface area of about 13.27 cm2 was exposed to the corrosive solution as shown in Figure 3.1. The mild steel have the following chemical analysis as obtained by Specialized Institute of Engineering Industries and shown in Table 3.3 below:-

62

Chapter Three

Experimental Work

Table 3.3 Chemical Composition of API X65 Mild Steel Alloy (wt %). Compound Al Cr Mo S V B Cu Nb

wt.% 0.032 0.011 0.103 0.004 0.055 0.0002 0.01 0.03

Compound Si C Ni Sn Ca Mn P Ti Fe

wt.% 0.24 0.15 0.02 0.005 0.0032 1.34 0.011 0.001 Reminder

3.1.5 Cleaning Materials: Analar benzene (C6H6 of 98.9 % purity) and analar acetone (C3H6O of 99.5 % purity) were used for cleaning and drying the samples.

3.2 Equipment and Accessories: 3.2.1 Equipment: 1. Potentiostat: is the electronic hardware required to control a three electrode cell and run most electrolytic experiments. It is equipped with current and voltage limiters offers better stability and control (type PRT 10-0.5). 2. Multi-Range Voltmeter: Multi range voltmeter was used to measure the potential, type AV08040B (made in China). 3. Mult-Range Ammeter: Multi range ammeter was used to measure the total current passing through the cell, type AV08040B (made in China). 4. Heater Magnetic Plate: Ordinary heater magnetic plate with thermostat was used, type Boeco MSH-300N (made in Germany). 5. Electronic Digital Balance: A digital balance with 4 decimal points was used, type Sartorius BP3015 (max. = 303 g, d = 0.1 mg).

63

Chapter Three

Experimental Work

6. Tecquiment Limited Nottingham: System rotation with a range of (03000 rpm) made in England. 7. pH-meter: A digital pH-meter with two decimal points to read hydrogen ion concentration (pH), type CHEMTRIX, it was calibrated before any measurement using standard buffer solutions of 2, 4, 7, 9 and10. 8. Dissolved Oxygen meter: A digital dissolved oxygen meter with two decimal points to read dissolved oxygen concentration with a range of (020 mg/l), type Lutron DO-5509 (made in Taiwan).

3.2.2 Accessories: 1. 2. 3. 4.

Different beakers Hooks Stand Pipette

5. 6. 7. 8.

Cylinder 10 milliliter & 5 milliliter. Connection wires. Epoxy. Thermometer.

3.3 Weight Loss Measurements: Weight loss technique is the widely used method for determining the corrosion rate. The corrosion of API X65 mild steel in CO 2 saturated, 3.5 wt. % NaCl solution was studied in presence and absence of acetic acid. 32 runs were carried out at desired different conditions. In addition to, 10 runs were performed in simulated brine and compared to 3.5 wt. % NaCl solutions saturated with CO2 gas in presence and absence of acetic acid at different experimental conditions. A typical procedure adopted was as follows: a specimen of known surface area and mass is exposed to the test corrosive solution for a fixed period of time 3 h. The loss of a metal as a result of corrosion is then determined from the loss of mass in specimen after removal the corrosion products or other deposits from metal. Mass loss values are usually recorded together with the exposed surface area of the specimen and the period of the test. Frequently the

64

Chapter Three

Experimental Work

data are expressed as mass loss per unit time per unit area, e.g. g/m2.day (gmd). In this work, cylinder shape specimen of API X65 mild steel alloy with dimensions 2.03 cm outside diameter, 2.08 cm long and 0.13 cm thickness, exposing a surface area of about 13.27 cm2 to corrosive media. Specimens were cleaned by washing with detergent and flushed with tap water followed by distilled water, degreased by analar benzene and acetone. Before each run, specimens of API X65 mild steel were abraded in sequence using emery paper of grade number 220, 320, 400 and 600, then washed with running tap water followed by distilled water then dried with clean tissue, degreased with benzene, dried, degreased with acetone, dried, and finally left in dissector over silica gel. To begin, the experimental apparatus was assembled, a salt solution was prepared, added to the glass cell, and then de-oxygenated for one hour using pure carbon dioxide gas ( > 99.99%). The test temperature was set using a hot plate and controlled by using a thermostat. Once deoxygenation had occurred and the test temperature was reached, the appropriate amount of HAc was then added to the cell and deoxygenation continued for an additional 30 minutes. Since HAc is volatile and the bubbling CO2 gas could strip the HAc out of the test cell, a preconditioning cell was used. The preconditioning cell was kept constant at the test temperature and contained the same fluid composition as the experimental cell. The preconditioning cell ensured the CO2 entering the experimental cell was saturated with HAc and H2O vapor by observing the concentration of dissolved oxygen continuously by a sensor probe (can be read directly to be less than 0.01 ppm on the display of the dissolved-oxygen meter). The pH meter used in the experiments was calibrated before any measurement using standard buffer solutions of 2, 4, 7, 9 and10. The pH was monitored before and after the HAc addition to 65

Chapter Three

Experimental Work

ensure the fluid composition was similar between test runs. In order to achieve the desired system pH, minute adjustments were made using droplets of hydrochloric acid and sodium bicarbonate solutions. After approximately 30 additional minutes, a pre-weighed the specimen was carried out using 4 decimals digital balance and its dimensions were measured with vernier, the electrode was fixed on the shaft, then immersed into 1000 cm3 test solution and the electrode’s speed of rotation was set and measurements started for a period of three hours. It is important to point out that the CO2 injection was maintained during the test with the aim of avoiding any oxygen contamination. During the three hours weight loss experiments, the pH was adjusted approximately every 2 minute for the first ten minutes, then every 5 min. for the remaining half hour until reaching the constant desired system pH (within ± 0.1) meaning that the system was in equilibrium and saturated with CO2. After three hours exposure period, the specimen was taken out from the test solution, washed with running tap water, scrubbed with a brush to remove corrosion products, dried, rinsed with ethanol and then weighed after drying. The experiments were repeated at least twice in order to ensure reasonable reproducibility. Figures 3.2 and 3.3 show, the system used in weight loss experiments. The Teflon shaft shown in Figure 3.2 was driven by a 0.25 h.p. vibrationless motor supplied by tecquipment limited (serial no. 061 TMI6), England. The lower part of the shaft carried one prepared cylindrical specimen which was exposed to the 3.5 wt. % NaCl solution saturated with CO2.

3.4 Design of Experimental Polarization Holder: (i) Teflon Shaft Description: The following materials were used to construct the shaft (holder) shown in Figure 3.4: 66

Chapter Three

Experimental Work

1. Teflon rod. 2. Brass rod. 3. Ball bearing. 4. Cylinder bearing. 5. Two different pieces of Teflon 4 cm in diameter. 6. Copper cylinder 2 cm long, 1.76 cm in diameter. 7. Two carbon brushes with 2 springs. 8. Teflon washer and Teflon cap.

(ii) Procedure: 1. The Teflon shaft 26.5 cm long and 1.75 cm outside diameter was made by machining. A hole 1.1 cm in diameter was drilled along the Teflon axis for the insertion of brass rod 29 cm long and 1.05 cm outside diameter which was made by machining also such that on one end the specimen was screwed and the other end was attached to the 0.25 h.p. motor. 2. Two holes 0.5 cm in diameter and 18 cm from the lower end of the shaft were drilled through the Teflon shaft body for the insertion of 2 springs 0.5 cm in diameter which were inserted through the holes to affect an electrical contact between the copper cylinder and potentiostat. 3. Two holes 4 and 2 cm in diameter was drilled through the two Teflon to a depth of one cm for insertion of the two ball and cylinder bearings. The first for supporting the shaft vertically through the polarization cell hole, and the second for supporting the shaft vertically through the holder bottom hole. 4. Further 2 holes each 0.6 cm were drilled through the upper Teflon mentioned in No.3 for inserting two-carbon brushes, so that the latter will be in contact with copper cylinder (which is in contact with brass rod ) from the outside. 67

Chapter Three

Experimental Work

5. The end of the carbon brushes were fixed on the upper Teflon body by small screws for electrical contact with the potentiostat.

3.5 Potentiodynamic Polarization Technique: Open circuit potential was carried out in a cell described in Figure 3.5 and polarization was carried out in a cell with electrical circuit described in Figure 3.6.

3.5.1 System Description: A total of 36 test runs were carried out in presence and absence of acetic acid at different experimental conditions using potentiodynamic polarization technique. Tests were carried out using a glass cell of 2.5 liter, as shown in Figure 3.6. The cell was equipped with eight holes, seven of them are used. One for the working electrode (mild steel cylinder). One had a cylindrical hole for mounting the lugging capillary prob., one for thermometer, one for the counter graphite electrode, one for pH probe and two for gas inlet and outlet. All potential values were measured in reference to saturated calomel electrode (SCE). The lugging capillary prob. was adjusted such that it was at a distance not more than 1 mm from the working electrode. The working electrode was (2.03 cm outside diameter x 2.08 cm long) API X65 mild steel alloy cylinder; this cylinder was fixed on brass zone on the shaft. Graphite electrode was used as a counter electrode has a dimension of (3.1cm. diameter x 4 cm. long), two wires were connected to a cylindrical concentric graphite electrode, and then fixed within Teflon ring. The electrode was mounted directly to the working electrode. Saturated calomel electrode (SCE) was used as reference electrode. To ensure that KCl solution was saturated, a small amount of KCl (solid) was kept in the solution of (SCE) as long as the test.

68

Chapter Three

Experimental Work

3.5.2 Preparation: 1. Specimens preparation was carried out as in weight loss technique described previously. 2. Washing the cell and electrodes with running tap water followed by distilled water and then by NaCl test solution. The cell then filled with 1000 cm3 test NaCl solution and then the cell was placed on the heater magnetic plate for a suitable time until reaching the thermal equilibrium. 3. De-oxygenation for one hour using carbon dioxide gas, the test temperature was reached, the appropriate amount of HAc was then added to the cell and de-oxygenation continued for an additional 30 minutes Yuli and Mokhtar, [2010]. 4. The preconditioning cell was kept constant at the test temperature and contained the same fluid composition, ensured the CO2 entering the experimental cell was saturated with HAc and H2O vapor by observing the concentration of dissolved oxygen was continuously detected by a sensor probe (can be read directly to be less than 0.01 ppm on the display of the dissolved-oxygen meter). 5. The pH was monitored before and after the HAc addition to ensure the fluid composition was similar between test runs. In order to achieve the desired system pH, minute adjustments were made using droplets of hydrochloric acid and sodium bicarbonate solutions. 6. The electrodes were fixed, immersed into the test solution, electrical connections were made, the electrode’s speed of rotation was set and measurements started.

3.5.3 Open Circuit Potential Measurement: The electric circuit described in Figure 3.5 represents the open circuit potential measurements in which the range in potential of the

69

Chapter Three

Experimental Work

working electrode was recorded as a function of time against saturated calomel electrode (SCE) bridged by a luggin Haber Probe. The testing method is accomplished by placing the luggin Haber capillary at a distance of 1 mm from the working electrode. The open circuit potential can be read directly on the display of the voltmeter.

3.5.4 Potentiodynamic Polarization Measurements: 1. After preparation of the working electrode, it was placed in position and immersed in the cell test solution. The corrosion cell parts were joined to each other, and then connected to the potentiostat, ammeter and voltmeter. 2. The reference and counter electrodes were fixed as mentioned. 3. The cathodic polarization is carried out beginning from highest negative potential of (-1000 mV) until reaching the corrosion potential. The potential was changed (10-15 mV) for each step after a one-minute period, the current is recorded. 4. The anodic polarization readings start at a potential resulting in a zero current density and is increased in a step of (10-15 mV) with recording of current at each step for one minute interval until a potential of about (-100 mV).

70

Chapter Three

Experimental Work

O.D. = 2.03 cm L = 2.08 cm

Fig. 3.1 Cylindrical Specimens of API X65 Mild Steel Alloy Used as Working Electrode.

71

Chapter Three

Experimental Work

0.25 h.p. AC/DC Motor

Ring Washer

Teflon Shaft

Thermometer

Ball Bearing

From CO2 Cylinder Electrode

CO2 Gas Outlet

Teflon Covers

Test Cell Dissolved O2 Gas Meter Electrode

pH Meter Electrode

CO2 Gas Distributer

Rotating Carbon Steel Cylindrical Specimen

Corrosive Solution Teflon Cap

Hot Plate Fig. 3.2 Experimental Set-up for Weight Loss Investigation.

72

Chapter Three

Experimental Work

Fig. 3.3 Photograph of Weight Loss System Set up.

73

Chapter Three

Experimental Work

0.25 h.p AC/DC Motor Motor

Brass Rod Brass Rod

Cu-Ring Teflon Shaft Carbon

Ring Washer

Spring Spring

Ball Bearing

Wire for Electrical Contact

Teflon Holder Body Cylinder Bearing

Teflon Holder Body Brass Rod

Teflon Shaft

Test Carbon Steel Cylindrical Specimen Teflon Cap

Fig. 3.4 Shaft Characterizations for Polarization Studies.

74

Chapter Three

Experimental Work

Fig. 3.5 Open Circuit Potential Measurment.

75

Chapter Three

Experimental Work

(1)

(7)

V

A

(8)

(10) (3)

(4)

(2)

(6)

CO2

(5)

(9)

Fig. 3.6 Electrochemical Polarization System Setup. (1) Potentiostat (2) Glass Test (3) Reference Electrode (4) Counter Electrode (5) Working Electrode (6) CO2 Gas Distributer (7) Ammeter (8) Voltmeter (9) Hot Plate (10) CO2 Gas Outlet

76

Chapter Three

Experimental Work

3.6 Tests for Protective and Non Protective Films: 3.6.1 Protective Film Thickness Test: The film thickness was calculated by using (Gravimetric determination of film weight and thickness) according to the standard test method (ASTM B 680-80) Frankel, [2004], from the weight loss measured in Equation (3.1). The specimens were weighed before immersion in the following chemical etching solution for mild steel Uhlig, [2013]: Hydrochloric acid

30 ml

Ferrous chloride

10 g

Deionized water

120 ml

The time for immersion is 10 min. and the temperature of electrolyte is 25°C. This solution does not attack mild steel but dissolves and cleans the thin corrosion products film (i.e., FeCO3 ) and leaving the surface of mild steel. The specimen is weighed before and after film removal. The loss in weight divided by the area provides a figure for the mass of film per unit area Marcus and Herbelin, [1993].

Where: = Weight of a specimen with the film in (g); = Weight of a specimen after film removal in (g); A = Surface area (dm2). The density of a 100% FeCO3 film is 3.96 g/cm3 Roberge, [2012]. From the weight loss measured above, the film thickness can be calculated from Equation (3.2):

77

Chapter Three

Experimental Work

where: = Film thickness in (micron). = Density in

⁄

.

A part from the appearance of the specimen after film removed, the completion of the stripping operation can be checked by repeating the immersion in chemical etching solution for another minute followed by drying and re-weighing. No further loss of weight shall occur. The specimens were rinsed in running water followed by distilled water to remove the excess solution, dried and weighed, the immersion in the above solution was repeating for another minutes followed by drying and re-weighing, no further loss in weight shall occur. The calculations of protective film thickness obtained in absence of acetic acid at the optimum conditions are listed in Table J.1 in Appendix J. The porosity calculations of the protective film obtained are described in Appendix J.

3.6.2 X-Ray Diffraction: To characterize and determine the phases of corrosion products present in the API X65 mild steel in presence and absence of acetic acid under the optimum conditions, X-ray diffraction studies were undertaken in (S.C. of Geological Survey and Mining) by using X-ray instrument type (Philips PW- 1840 (made in Holland)) X-ray diffractometer with the setting listed in Table 3.4.

78

Chapter Three

Experimental Work

Table 3.4 X-Ray Diffractometer Settings. Scan Speed Wave Length Current Voltage

5 deg/min λ = 1.54060 Å 30 mA 40 kV

The X-ray examination was carried out to make sure that the entire surface was converted to scale film of corrosion products (iron carbide and ferrous carbonate) in presence and absence of acetic acid. In this work, ring shape specimen of API X65 mild steel with dimension 2.03cm outside diameter, 2.08 cm width and 0.13 cm thickness. The peak of each element is aimed to describe the chemical compound formed of the corroded steel surface. Peak identification was made on the basis of d spacing of typical compounds given in the ASTM powder diffraction files of Ministry of Industry and Minerals-Baghdad. The keys used to identify the peaks in the X-ray diffraction, which were obtained in the work are given in Tables J.2 and J.3 in Appendix J.

3.6.3 Roughness Test: Measuring the surface roughness of the protective mild steel film was carried out using (Taysurf – Taylor-Hobson Company- England) shown in Figure 3.7 with the following specifications: Stylus material: Diamond Stylus force: 100 mg.f The arithmetic average values (Ra) for no protective and protective film formation specimens were calculated directly from the instrument. The examinations were carried out for API X65 mild steel with absence and presence of protective film formation specimens.

79

Chapter Three

Experimental Work

Fig. 3.7 Taysurf System for Measuring Surface Roughness.

3.6.4 Hardness Test: Vickers micro hardness (VMH) test was carried out for the specimens under optimum conditions for API X65 mild steel under no protective

and

Computerized

protective Metallurgical

film

formation

Optical

by

Microscopy

using

LARYEE

Technique

(W.

Germany) as shown in Figure 3.8. The magnification was X100 and the applied load was 9.8 N. The average of 5 readings of the indentation length was taken while; the applied load was kept for 20 sec for each reading. The Vickers microhardness (VMH) was computerized according to the following Equation:

where: F: The force (load) = 9.8 N D: The diameter of measuring point in (µm).

80

Chapter Three

Experimental Work

Fig. 3.8 Vicker Micro Hardness System for Measuring Surface Hardness.

3.6.5 Microstructure Examination: The surface and cross-section microstructures evolution were characterized by using Scanning Electron Microscopy (SEM) and a Computerized Metallurgical Optical Microscopy Technique (CMOMT), (Type MeF2), (Carllnsize Company - (Instruments Analytical, W. Germany, for analysis of the surface films)) with digital camera. The films were examined visually by optical microscopy under (5X) magnification power. These tests were achieved at the Ministry of Science and Technology-Baghdad. The specimens were prepared from API X65 mild steel before and also after corrosion in CO 2 saturated,3.5 wt % NaCl solution in presence and absence of acetic acid at the optimum conditions by the weight loss measurement (WL) for a period 3 h to characterize and note the non protective and protective film formation.

81

Chapter Four Results & Discussion

Chapter Four

Results & Discussion

T

he corrosion behavior of API X65 mild steel in CO2 saturated, 3.5 wt % NaCl solution in presence and absence of acetic acid at

different experimental conditions have been investigated by weight loss and polarization techniques as shown in Table 4.1 below: Table 4.1 Different Experimental Conditions of Parameters. Variable Temperature (°C) pH HAc Acid Conc. (ppm) Speed of Rotation (rpm)

No Protective Film Formation (Presence of HAc) 40, 50 & 60 °C 3, 4 & 5

Presence of Protective Film Formation (Absence of HAc) 65, 70 & 75 °C 7.5, 8 & 8.5

1000, 2000 & 3000

----

1000, 1250 & 1500

1000, 1250 & 1500

4.1 Weight Loss Measurements: 4.1.1 Experiments in a Simulated Brine Solution: A series of weight loss technique experiments were performed in simulated brine compared to 3.5 wt% NaCl solutions saturated with CO2 gas in presence and absence of acetic acid at different experimental conditions to study and verify the effect of the presence of multiple ions (Na+, Cl-, K+, Ca+2, Mg+2) on the corrosion rates of API X65 mild steel. It was found that no significant difference can be seen from the weight loss measurements over three hours between the two solutions is shown in Appendix K.

4.1.2 Experimental Strategy in NaCl Solutions: 4.1.3 No Protective Film Formation: The corrosion rates of API X65 mild steel in 3.5 wt % NaCl saturated with CO2 solutions in presence of different concentrations of

82

Chapter Four


acetic acid and different temperatures, pH’s with speeds of rotation are summarized in Table 4.2 through 20 runs using weight loss technique. Table 4.2 Weight Loss Corrosion Rates (gmd) Results in Presence of Acetic Acid (Absence of Protective Film Formation). Run No.

T (ºC)

pH

CA (ppm)

ω (rpm)

W1 (gm)

W2 (gm)

C.R (gmd)

1.

40

3

1000

1000

25.4825

25.4744

48.52

2. 3. 4. 5. 6. 7. 8. 9. 10.

60 40 60 40 60 40 60 40 60

3 5 5 3 3 5 5 3 3

1000 1000 1000 3000 3000 3000 3000 1000 1000

1000 1000 1000 1000 1000 1000 1000 1500 1500

25.3988 25.1123 25.0431 24.7991 24.6995 24.7881 25.1199 24.2213 25.2182

25.3788 25.1057 25.0349 24.7907 24.6783 24.7811 25.1106 24.2127 25.1975

121.30 40.06 50.19 51.22 129.39 42.53 56.14 52.65 125.15

11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

40 60 40 60 40 60 50 50 50 50

5 5 3 3 5 5 4 4 4 4

1000 1000 3000 3000 3000 3000 2000 2000 2000 2000

1500 1500 1500 1500 1500 1500 1250 1250 1250 1250

24.3171 25.3915 24.6933 24.3487 24.7913 24.5321 25.3923 24.5318 24.7312 24.7110

24.3099 25.3831 24.6841 24.3272 24.7837 24.5221 25.3817 24.5219 24.7208 24.7013

43.67 51.14 55.69 130.46 46.02 60.14 63.81 59.72 62.43 58.64

4.1.3.1 Study Area: The field of variation of the 4 studied experimental factors was selected in order to approach the natural conditions which can be met in the experimental field. Classically, the various levels were expressed in a system of coded variables. Level +1 corresponded to the highest real value and level -1 to the lowest real value. The correspondence between real variables and coded ones was done starting from the following Equation: (

)

( 83

)

Chapter Four

Where:



value of factor i in real variable; and

= corresponding


variation and k is the number of the input factors: (

)

(

)

All fields of variation for the 4 studied factors are warranted in Table 4.3: Table 4.3 Center and Variation Step of Parameters. X1 X2 X3 X4

Factor Temperature pH HAc Acid Concentration Speed of Rotation

Unit °C ppm rpm

Centre 50 4 2000 1250

Variation Step 10 1 1000 250

4.1.3.2 Experimental Response: The only experimental response followed in the current study (weight loss technique) was the corrosion rate in (gm/m2.day); it is calculated by using the following Equation: ( Where

and

)

(

)

(

)

are the weights of mild steel before and after weight

loss in grams, respectively, A: surface area and t: time in day.

4.1.3.3 Used Matrix: The response of experiments conducted according to a four-factors, two-levels Full Factorial Experimental Design (FFED) method are represented by corrosion rate to evaluate statistically the effects of each factor over the specified range studied and the interactions among the effects Box and Hunto, [2005]. The results of experimental runs are shown in Table 4.2, and a sample of calculation is shown in Appendix A.

84

Chapter Four


Table 4.4 shows the results of corrosion rates of the 20 experiments conducted at low (-1), center (0) & high (+1) levels of the studied variables. The selected design matrix was a Full Factorial Experimental Design (FFED) consisting of 20 rows of coded/real factors, corresponding to a number of trials. This design provides a uniform distribution of experimental points within the selected experimental hyper-space and the experiment with high resolution. The corrosion rate values (C.R), shown in Table 4.4, are the values obtained by the weight loss measurement technique. Table 4.4 Coded & Real Variables with the Observed Values of the Response, (C.R) in Presence of Acetic Acid (Absence of Protective Film Formation). Run No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

T 40 60 40 60 40 60 40 60 40 60 40 60 40 60 40 60 50 50 50 50

Real factor pH CA 3 1000 3 1000 5 1000 5 1000 3 3000 3 3000 5 3000 5 3000 3 1000 3 1000 5 1000 5 1000 3 3000 3 3000 5 3000 5 3000 4 2000 4 2000 4 2000 4 2000

ω 1000 1000 1000 1000 1000 1000 1000 1000 1500 1500 1500 1500 1500 1500 1500 1500 1250 1250 1250 1250

x1 -1 +1 -1 +1 -1 +1 -1 +1 -1 +1 -1 +1 -1 +1 -1 +1 0 0 0 0

85

Coded factor x2 x3 -1 -1 -1 -1 +1 -1 +1 -1 -1 +1 -1 +1 +1 +1 +1 +1 -1 -1 -1 -1 +1 -1 +1 -1 -1 +1 -1 +1 +1 +1 +1 +1 0 0 0 0 0 0 0 0

x4 -1 -1 -1 -1 -1 -1 -1 -1 +1 +1 +1 +1 +1 +1 +1 +1 0 0 0 0

Response C.R (gmd) 48.52 121.30 40.06 50.19 51.22 129.39 42.53 56.14 52.65 125.15 43.67 51.14 55.69 130.46 46.02 60.14 63.81 59.72 62.43 58.64

Chapter Four


4.1.3.4 Statistical Treatment of Data: The linear, the quasi-linear, the quadratic, the power non-linear and the exponential & power mathematical models were selected for the analysis in this study. The parameters of Equations (2.75 through 2.81) have been estimated by means of the least-square method, using STATISTICA program software package® v.10. In this way, the following multiple regression Equations in presence of acetic acid (absence of protective film formation) were obtained: 1) ̂

(

)

(

)

(

)

(

)

(

)

2) ̂

3) ̂

4) ̂ 5) ̂

The coding of the process factors was carried out according to the Equation (2.73a). Where xi, i = 1, 2, 3 &4 are in their coded levels ranged (-1, 0, +1), then Equation (2.73a) is applied to setup the relationship between the coded level and the corresponding real variables.

The fitted

multiple regression Equations in terms of the real levels of the solution temperature, the pH, the HAc acid concentration and the speed of rotation may be obtained by substituting the transforming Equation (2.73a) into the Equations (4.2 through 4.6) as follows: 86

Chapter Four


From the regression Equation (the quadratic model) the optimized values are calculated by partial differentiating the above Equation with respect to x1, x2, x3 & x4 and equating to zero solving the four Equations from partial differentiation results in the optimal corrosion rate. The final optimal conditions are: Variable Temperature, °C pH HAc Acid Conc., ppm Speed of Rotation, rpm

Code -0.5 0.8 0.2 0.2

Real 45.4 4.8 2178.5 1296.6

Figures 4.1 through 4.4 clearly show that the quadratic mathematical model (R = 0.999) most accurately approximates the experimental results in presence of acetic acid (absence of the protective film formation). 80 75

linear model ( R = 0.877 ) quadratic model ( R = 0.999 ) quasi-linear model ( R = 0.994 ) power (nonlinear) model ( R = 0.966 ) exponential & power model ( R = 0.965 )

Corrosion Rate (gmd)

70 65 60 55 50 45 40 35 30 25 38

40

42

44

46

48

50

52

54

56

58

60

62

Temperature ( oC)

Fig. 4.1 Relationship between the Corrosion Rate and the Temperature for Different Regression Equations at the Optimum Conditions (pH 4.8, 2178.5 ppm & 1296.6 rpm) without Protective Film Formation. 87

Chapter Four

Results & Discussion 85 80



75 70 65 60 55 50 45 40 35 2.8

3.0

3.2

3.4

3.6

3.8

4.0

4.2

4.4

4.6

4.8

5.0

5.2

pH

Fig. 4.2 Relationship between the Corrosion Rate and the Acidic Function for Different Regression Equations at the Optimum Conditions (45.4 °C, 2178.5 ppm & 1296.6 rpm) without Protective Film Formation.

50


48


46

44

42

40

38 800

1000

1200

1400

1600

1800

2000

2200

2400

2600

2800

3000

3200

HAc Concentration (ppm)

Fig. 4.3 Relationship between the Corrosion Rate and the Acetic Acid Concentration for Different Regression Equations at the Optimum Conditions (45.4 °C, pH 4.8 & 1296.6 rpm) without Protective Film Formation.

88

Chapter Four

Results & Discussion 50


48


46

44

42

40

38 900

1000

1100

1200

1300

1400

1500

1600

S peed of Rotation (rpm)

Fig. 4.4 Relationship between the Corrosion Rate and the Speed of Rotation for Different Regression Equations at the Optimum Conditions (45.4 °C, pH 4.8 & 2178.5 ppm) without Protective Film Formation.

4.1.3.5 Validity of the Model: To determine the significant of the above mentioned effect, an analysis of variance (ANOVA) was carried out. The corresponding analysis of variance is represented in Tale 4.5 for corrosion of API X65 steel in CO2 saturated, 3.5 wt. % NaCl solution in presence of acetic acid (absence of the protective film formation) using Full Factorial Experimental Design (FFED) methodology for non-linear (quadratic) mathematical model with the regression coefficient of the model is 99.99 %. The result obtained from this analysis indicates the significance of variables studied through the Student F-value at 95% confidence level. Calculation of statistics data used STATISTICA program software package® version 10. See Appendix C.

89

Chapter Four


Table 4.5 Analysis of Variance (ANOVA) for API X65 Mild Steel Alloy Corrosion in Absence of Protective Film Formation.

Linear Square

b1 b2 b3 b4 b11 b22 b33 b44 b12 b13 b14 b23 b24 b34

Source

Interaction

Constant Estimated

∑ x2 16 16 16 16 16 16 16 16 16 16 16 16 16 16

Estimate Coefficient (b) 21.4734 -20.2808 2.4315 1.5982 1.9664 1.9664 1.9664 1.9664 -15.8045 1.1114 -0.3647 0.0406 -0.0914 0.0309

Variance Sb2=Sr2/ ∑ x2 0.2735 0.2735 0.2735 0.2735 0.2735 0.2735 0.2735 0.2735 0.2735 0.2735 0.2735 0.2735 0.2735 0.2735

F-value = b2/Sb2

F0.95(1,5) = 6.61

1685.9485 1503.8788 21.6168 9.3391 14.1379 14.1379 14.1379 14.1379 913.2805 4.5163 0.4863 0.0060 0.0305 0.0035

S S S S S S S S S NS NS NS NS NS

The new response function is then written in the following form: In Absence of Protective Film Formation: ̂ (

)

The preliminary information of the quantitative and qualitative impact on the objective function (response) of each individual factor in the regression Equations can be obtained from its parameters sign and magnitude Jeff Wu and Michael, [2009]. The positive sign for the parameters of the temperature of solution, the HAc acid concentration and the speed of rotation indicates that the corrosion rate increases (response surface deteriorates) with the increase in these three factors. The negative sign for the parameter of the solution pH shows that the corrosion rate decreases (response surface improves) with the increase in the solution pH. Furthermore, the given quadratic regression Equation and Pareto 90

Chapter Four


chart as shown in Figure 4.5 suggest that the dominant process factor is the temperature of solution, while the effects of the solution pH, the HAc acid concentration and the speed of rotation are considerably smaller. The factor interactions and quadratic (square) have generally the least influence on the considered problem, except the factor interaction between the temperature and the solution pH. In order to take into account the contribution from the remainder factor interactions, these terms were not neglected. T pH T x pH CA T2 pH 2 CA2 ω2 ω T x CA Txω pH x ω pH x CA CA x ω

Quadratic in Code

0

2

4

6

8

10

12

14

16

18

20

22

24

Fig. 4.5 Pareto Chart for API X65 Mild Steel in Presence of Acetic Acid (Absence of Protective Film Formation). The criterion used to estimate the efficiency and ability of the mathematical model to predict corrosion rate could be the absolute percentage error - ‫ ׀‬δi ‫׀‬, which is defined by Equation: | | where

|

| and

( )

(

)

represent the predicted and measured corrosion rate

for i-th trial, respectively. The corrosion rate calculated according to the Equation (4.7), and errors calculated according to the Equation (4.8) are given in Table 4.6.

91

Chapter Four


Table 4.6 Experimental, Predicted Results and Absolute Percentage Error for API X65 Mild Steel in Presence of Acetic Acid (Absence of Protective Film Formation). Coded factor

Response (gmd)

Run No.

x1

x2

x3

x4

C.R

Exp. Error ̂

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

-1 +1 -1 +1 -1 +1 -1 +1 -1 +1 -1 +1 -1 +1 -1 +1 0 0 0 0

-1 -1 +1 +1 -1 -1 +1 +1 -1 -1 +1 +1 -1 -1 +1 +1 0 0 0 0

-1 -1 -1 -1 +1 +1 +1 +1 -1 -1 -1 -1 +1 +1 +1 +1 0 0 0 0

-1 -1 -1 -1 -1 -1 -1 -1 +1 +1 +1 +1 +1 +1 +1 +1 0 0 0 0

48.52 121.30 40.06 50.19 51.22 129.39 42.53 56.14 52.65 125.15 43.67 51.14 55.69 130.46 46.02 60.14 63.81 59.72 62.43 58.64

-0.20 -0.48 0.19 0.48 0.01 0.67 0.00 -0.68 -0.11 0.78 0.12 -0.79 0.31 -0.97 -0.31 0.98 2.66 -1.43 1.28 -2.51

48.72 121.78 39.87 49.71 51.21 128.72 42.53 56.82 52.76 124.37 43.55 51.93 55.38 131.43 46.33 59.16 61.15 61.15 61.15 61.15 The mean absolute percentage error | ̅ |

Error | | 0.41 0.39 0.47 0.96 0.02 0.52 0.00 1.21 0.21 0.62 0.27 1.54 0.56 0.74 0.67 1.63 4.17 2.39 2.05 4.28

The accuracy of any empirical model can also be done by means of statistical

parameters,

for

example,

correlation

coefficient.

The

correlation coefficient (R) is a statistical measure of the strength of correlation between the predicted and measured values Devore, [2005]. For the current problem, the following result is obtained: R = 0.999 in presence of acetic acid (absence of the protective film formation) as shown in Figure 4.6.

92

Chapter Four


C.R Predicted (gmd)

120

100

80

60

40

Experimental Data Points Quadratic Model (R = 0.999)

20 20

40

60

80

100

120

140

C.R Measured (gmd)

Fig. 4.6 Performance of the Quadratic Mathematical Model for API X65 Mild Steel in Presence of Acetic Acid (Absence of Protective Film Formation).

4.1.3.6 Graphic Analysis of the Model: The aim of this study was to find a corrosion rate whose features would have been previously defined from the operative conditions extracted from the quadratic mathematical model. Because the direct exploitation of the Equation was delicate, it was convenient to restore it under a graphic representation; while fixing two of the four factors of the survey, it was possible to represent the response surface materializing the surface of regression in a three-dimensional space. It was also possible to project the Equation in a design under isoresponse curves, interpreted as card curves level. (i) Evolution of Corrosion Rate as a Function of the Temperature and the pH: Figure 4.7 shows the evolution of the corrosion rate as a function of the temperature and the pH. It can be seen that the temperature has a strong influence on the tentative response. The minimal corrosion rate is obtained for a temperature of -0.5 in coded variable, i.e., 45.4 °C in real variable. Considering simultaneous effects of temperature and pH is 93

Chapter Four


presented in Figure (4.7 contour plot). The Figure shows, in low pH (pH 3), the increase of corrosion rate is higher than in higher pH (pH 5).

180 160

) Corrosion Rate (gmd

140 1 20 100 80 60 5.2

5.0

4.8

4.6

4.4

4.2

pH

4.0

3.8

3.6

3. 4

3. 2

3. 0

2. 8

38

40

44

42

46

48

Te

50

52

54

58

56

0

e( tur e ra p m

60

62

C) > 160 < 152 < 132 < 112 < 92 < 72 < 52

3D Contour Plot 5.2 5.0 4.8


4.6 4.4

pH

4.2 4.0 3.8 3.6 3.4 3.2 3.0 2.8 38

40

42

44

46

48

50

52

54 o

Temperature ( C)

56

58

60

62

152 132 112 92 72 52

Fig. 4.7 Response Surface Plot (top) and Contour Plot (bottom) Showing the Variation of Response (Corrosion Rate) as a Function of the Temperature (X1) and pH (X2) at the Optimum Conditions (2178.5 ppm HAc & 1296.6 rpm).

94

Chapter Four


(ii) Evolution of Corrosion Rate as a Function of the pH and HAc Acid Concentration: Figure 4.8 shows the synergism between the two factors: the pH and HAc acid concentration in corrosion rate at temperature 45.4°C and speed of rotation of 1296.6 rpm. It can be noted that the effect of the HAc acid concentration differed according to the corrosion rate’s variation. This effect becomes positive and even more important when the corrosion rate is degraded (reduced). Analysis of corrosion rate as effects of interaction between pH and HAc acetic concentration is shown in Figure (4.8 contour plot). The model shows an increase of corrosion rate due to HAc. And the decrease of corrosion rate is caused by pH.

95

Chapter Four


90

80


70

60

50

40

00 32 00 30

28

00 00 26 00 24 00 22 00 20 00 18 00 16

Co HA nc c (pp entr Acid m) atio n

00 14 00 12 00 10 0 80

2.8

3.0

3.2

3.4

3.6

3.8

4.0

4.2

4.4

4.6

4. 8

5. 0

5. 2

pH

> 80 < 76 < 66 < 56 < 46 < 36

3D Contour Plot 5.2 5.0 4.8


4.6 4.4

pH

4.2 4.0 3.8 3.6 3.4 3.2 3.0 2.8 800

1200 1000

1600 1400

2000 1800

2400 2200

2800 2600


3200 3000

76 66 56 46 36

Fig. 4.8 Response Surface Plot (top) and Contour Plot (bottom) Showing the Variation of Response (Corrosion Rate) as a Function of the pH(X2) and HAc Acid Concentration (X3) at the Optimum Conditions (45.4 °C & 1296.6 rpm).

96

Chapter Four


(iii) Evolution of Corrosion Rate as a Function of the Temperature and Speed of Rotation: Figure 4.9 represent the evolution of the corrosion rate as a function of the temperature and the speed of rotation. This Figure shows that the corrosion rate decreased when the temperature decreased. This evolution was however more accentuated for the lower speeds of rotation. Considering simultaneous effects of temperature and speed of rotation is presented in Figure (4.9 contour plot). The Figure shows, in high speed of rotation (1500 rpm), the increase of corrosion rate is higher than in lower speed of rotation (1000 rpm).

97

Chapter Four


80 75


70 65 60 55 50 45 40 62

60

58

56

54 Te mp 52 era 50 48 tur e ( o 46 44 C) 42

40

38

90 0

10 00

11 00

12 00

13 00

o eed Sp

fR

14 00

o ta

tio

15 00

16 00

> < < < < < < <
48 < 46 < 42 < 38 < 34 < 30 < 26

3D Contour Plot 1600

1400


Speed of Rotation (rpm)

1500

1300

1200

1100

1000

900 800

1200 1000

1600 1400

2000 1800

2400 2200

2800 2600


3200 3000

46 42 38 34 30 26

Fig. 4.10 Response Surface Plot (top) and Contour Plot (bottom) Showing the Variation of Response (Corrosion Rate) as a Function of the HAc Acid Concentration (X3) and Speed of Rotation (X4) at the Optimum Conditions (45.4°C & pH 4.8). 100

Chapter Four


Similar results were obtained by other researchers Yuli and Mokhtar, [2010] and George and Nesic, [2004, 2007], for similar CO2 corrosion conditions without protective film formation. The optimization problem for the given corrosion process of API X65 mild steel in CO2 saturated, 3.5 wt% NaCl solution in presence of acetic acid under no protective film formation can be mathematically stated as follows: find the vector of CO2 corrosion factors (

),

(

), Equation (4.4) subjected to

which

minimizes

the

objective

function

̂

.

In this particular case, Equations (2.83) and (2.84) give in coded form optimum values and response: (

) ̂

According to the above described procedure, translating the coded in the real factors gives: ( ̂

̂

) 94 gmd

By using Equation (2.86), the following eigenvalues are obtained: ( ( (

)( )(

) )

).

These

eigenvalues satisfy Equation (2.87). Also, indicates that the considered system is really a saddle system, see Appendix D.

4.1.4 In Presence of Protective Film: The corrosion rates of API X65 mild steel in 3.5 wt % NaCl saturated with CO2 solutions in absence of acetic acid at different temperatures, pH’s with speeds of rotation are summarized in Table 4.7 through 12 runs using weight loss technique.

101

Chapter Four


Table 4.7 Weight Loss Corrosion Rates (gmd) Results in Absence of Acetic Acid (Presence of Protective Film Formation). Run No.

T (ºC)

pH

ω (rpm)

W1 (g)

W2 (g)

C.R (gmd)

1.

65

7.5

1000

25.2763

25.2737

15.88

2.

75

7.5

1000

24.6846

24.6788

35.49

3.

65

8.5

1000

24.6801

24.6781

12.44

4.

75

8.5

1000

24.5746

24.5694

31.82

5.

65

7.5

1500

24.1130

24.1097

20.53

6.

75

7.5

1500

25.5203

25.5143

37.09

7.

65

8.5

1500

25.2609

25.2582

16.35

8.

75

8.5

1500

25.0484

25.0431

32.76

9.

70

8

1250

24.8659

24.8628

19.17

10. 11. 12.

70 70 70

8 8 8

1250 1250 1250

24.5417 24.5678 24.4456

24.5386 24.5647 24.4425

19.20 18.99 18.95

4.1.4.1 Study Area: The field of variation of the 3 studied experimental factors was selected in order to approach the real conditions which can be met in the experimental field. Classically, the various levels were expressed in a system of coded variables. Level +1 corresponded to the highest real value and level -1 to the lowest real value. The correspondence between real variables and coded ones was done starting from the following Equation: ( Where:

)

(


value of factor i in real variable; and

)

= corresponding


variation and k is the number of the input factors: ( 102

)

(

)

Chapter Four


All fields of variation for the 3 studied factors are warranted in Table 4.8: Table 4.8 Center and Variation Step of Parameters. Factor Temperature pH Speed of Rotation

X1 X2 X3

Unit °C rpm

Centre 70 8 1250

Variation Step 5 0.5 250

4.1.4.2 Experimental Response: The only experimental response followed in the current study (weight loss technique) was the corrosion rate in (gm/m2.day); it is calculated by using the following Equation: ( Where

and

)

(

)

(

)

are the weights of mild steel before and after weight

loss in grams, respectively, A: surface area and t: time in day.

4.1.4.3 Used Matrix: The response of experiments conducted according to a threefactors, two-levels Full Factorial Experimental Design (FFED) method are represented by corrosion rate to evaluate statistically effects of each factor over the specified range studied and the interactions among the effects Box and Hunto, [2005]. The results of experimental runs are shown in Table 4.7, and a sample of calculation is shown in Appendix A. Table 4.9 shows the results for corrosion rate of the 12 experiments conducted at low (-1), center (0) & high (+1) levels of the studied variables. The selected design matrix was a Full Factorial Experimental Design (FFED) consisting of 12 rows of coded/real factors, corresponding to a number of trials. This design provides a uniform distribution of experimental points within the selected experimental 103

Chapter Four


hyper-space and the experiment with high resolution. The corrosion rate values (C.R), shown in Table 4.9, are the values obtained by the weight loss measurement technique. Table 4.9 Full Factorial Experimental Design of the Independent Variables with the Observed Values for the Response, (C.R) in Absence of Acetic Acid (Presence of Protective Film Formation). Real factor

Coded factor

Run No.

T

pH

ω

x1

x2

x4

1 2 3 4 5 6 7 8 9 10 11 12

65 75 65 75 65 75 65 75 70 70 70 70

7.5 7.5 8.5 8.5 7.5 7.5 8.5 8.5 8 8 8 8

1000 1000 1000 1000 1500 1500 1500 1500 1250 1250 1250 1250

-1 +1 -1 +1 -1 +1 -1 +1 0 0 0 0

-1 -1 +1 +1 -1 -1 +1 +1 0 0 0 0

-1 -1 -1 -1 +1 +1 +1 +1 0 0 0 0

Response C.R (gmd) 15.88 35.49 12.44 31.82 20.53 37.09 16.35 32.76 19.17 19.20 18.99 18.95

4.1.4.4 Statistical Treatment of Data: The linear, the quasi-linear, the quadratic, the power non-linear and the exponential & power mathematical models were selected for the analysis in this study. The parameters of Equations (2.75 through 2.81) have been estimated by means of the least-square method, using STATISTICA program software package® v. 10. In this way, the following multiple regression Equations in absence of acetic acid (presence of protective film formation) were obtained: (

1) ̂ 2) ̂

( 104

) )

Chapter Four


3) ̂ (

)

4) ̂

(

)

5) ̂

(

)

The coding of the process factors was carried out according to the Equation (2.73a). Where xi, i = 1, 2 & 3 are in their coded levels ranged (1, 0, +1), then Equation (2.73a) is applied to setup the relationship between the coded level and the corresponding real variables.

The fitted

multiple regression Equations in terms of the real levels of the solution temperature, the pH and the speed of rotation may be obtained by substituting the transforming Equation (2.73a) into the Equations (4.9 through 4.13) as follows:

From the regression Equation (the quadratic model) the optimized values are calculated by partial differentiating the above Equation with respect to x1, x2 & x3 and equating to zero solving the three Equations from partial differentiation results in the optimal corrosion rate. The final optimal conditions are: Variable Temperature, °C pH Speed of Rotation, rpm

Code -0.3 -0.2 0.7

Real 68.7 7.9 1425.8

Figures (4.11 through 4.13) clearly show that the quadratic mathematical model (R = 0.999) most accurately approximates the experimental results in absence of acetic acid (presence of protective film formation).

105

Chapter Four


45

linear model (R = 0.931) quasi-linear model (R=0.933) quadratic model (R=0.999) power (nonlinear) model ( R = 0.954 ) exponential & power model ( R = 0.948 )


40

35

30

25

20

15

10 64

66

68

70

72

74

76

o

Temperature ( C)

Fig. 4.11 Relationship between the Corrosion Rate and the Temperature for Different Regression Equations at the Optimum Conditions (pH 7.9 and 1425.8 rpm) with Protective Film Formation.

26 24


22 20 18 16 14

linear model (R = 0.931) quasi-linear model (R=0.933) quadratic model (R=0.999) power (nonlinear) model ( R = 0.954 ) exponential & power model ( R = 0.948 )

12 10 8 6 7.4

7.6

7.8

8.0

8.2

8.4

8.6

pH

Fig. 4.12 Relationship Between the Corrosion Rate and the pH for Different Regression Equations at the Optimum Conditions (68.7 °C and 1425.8 rpm) with Protective Film Formation.

106

Chapter Four

Results & Discussion 24 23


22 21 20 19 18 17

linear model (R = 0.931) quasi-linear model (R = 0.933) quadratic model (R = 0.999) power (nonlinear) model ( R = 0.954 ) exponential & power model ( R = 0.948 )

16 15 14 900

1000

1100

1200

1300

1400

1500

1600


Fig. 4.13 Relationship Between the Corrosion Rate and the Speed of Rotation for Different Regression Equations at the Optimum Conditions (68.7 °C and pH 7.9) with Protective Film Formation.

4.1.4.5 Validity of the Model: To determine the significant of the above mentioned effect, an analysis of variance (ANOVA) was carried out. The corresponding analysis of variance is represented in Table 4.10 for corrosion of API X65 steel in CO2 saturated, 3.5 wt% NaCl solution in absence of acetic acid (presence of the protective film formation) using Full Factorial Experimental Design (FFED) methodology for non-linear (quadratic) mathematical model with the regression coefficient of the model is 99.99 %. The result obtained from this analysis indicates the significance of variables studied through the Student F-value at 95% confidence level. Calculation of statistics data used STATISTICA program software package® version 10. See Appendix C.

107

Chapter Four


Table 4.10 Analysis of Variance (ANOVA) for API X65 Mild Steel Alloy Corrosion in Presence of Protective Film Formation.

b12 b13 b23

8

-0.0469

8 8

Linear

8 8 8 8 8 8

Variance Sb2=Sr2/ ∑ x2 0.0568 0.0568 0.0568 0.0568 0.0568 0.0568

∑ x2

Square

b1 b2 b3 b11 b22 b33

Estimate Coefficient (b) 8.9932 -1.9540 -1.3881 2.0725 2.0725 2.0725

Source

Interaction

Constant Estimated

F-value = b2/Sb2

F0.95(1,2) =10.13

1424.859 67.2673 33.9461 75.6716 75.6716 75.6716

S S S S S S

0.0568

0.0389

NS

-0.7521

0.0568

9.9663

NS

-0.1758

0.0568

0.5442

NS

The new response function is then written in the following form: In Presence of Protective Film Formation: ̂ (

)

The preliminary information of the quantitative and qualitative impact on the objective function (response) of each individual factor in the regression Equations can be obtained from its parameters sign and magnitude Jeff Wu and Michael, [2009]. The positive sign for the parameters of the temperature of solution and the speed of rotation indicates that the corrosion rate increases (response surface deteriorates) with the increase in these two factors. The negative sign for the parameter of the solution pH shows that the corrosion rate decreases (response surface improves) with the increase in the solution pH. Furthermore, the given quadratic regression Equation and Pareto chart as shown in Figure 4.14 suggest that the dominant process factor is the temperature of solution, while the effects of the solution pH and the speed of rotation are considerably smaller. The factor interactions have the least influence, but 108

Chapter Four


the factor quadratics (squares) has the equal influence on the considered problem. In order to take into account the contribution from the factor interactions, these terms were not neglected. TT 2 TT2 2 pH pH2 2 ωω2

pH pH

ωω T xT xωω pHpHx xωω T

Quadratic in Code

x pH xT pH 0

1

2

3

4

5

6

7

8

9

10

Fig. 4.14 Pareto Chart for API X65 Mild Steel in Absence of Acetic Acid (Presence of Protective Film Formation) The criterion used to estimate the efficiency and ability of the mathematical model to predict corrosion rate could be the absolute percentage error- ‫ ׀‬δi ‫׀‬, which is defined by Equation: | | where

|

| and

( )

( ) represent the predicted and measured corrosion rate

for i-th trial, respectively. The corrosion rate calculated according to the Equation (4.14), and errors calculated according to the Equation (4.8) are given in Table 4.11.

109

Chapter Four


Table 4.11 Experimental and Predicted Results and Absolute Percentage Error for API X65 Mild Steel in Absence of Acetic Acid (Presence of Protective Film Formation). Coded factor

Response (gmd)

Run No.

x1

x2

x3

C.R

1 2 3 4 5 6 7 8 9 10 11 12

-1 +1 -1 +1 -1 +1 -1 +1 0 0 0 0

-1 -1 +1 +1 -1 -1 +1 +1 0 0 0 0

-1 -1 -1 -1 +1 +1 +1 +1 0 0 0 0

15.88 35.49 12.44 31.82 20.53 37.09 16.35 32.76 19.17 19.20 18.99 18.95

Exp. Error ̂

15.89 35.48 12.43 31.83 20.52 37.10 16.36 32.75 19.08 19.08 19.08 19.08

-0.01 0.01 0.01 -0.01 0.01 -0.01 -0.01 0.01 0.09 0.12 -0.09 -0.13

Error | | 0.06 0.03 0.08 0.03 0.05 0.03 0.06 0.03 0.47 0.63 0.47 0.69

The mean absolute percentage error | ̅ |

The accuracy of any empirical model can also be done by means of statistical

parameters,

for

example,

correlation

coefficient.

The

correlation coefficient (R) is a statistical measure of the strength of correlation between the predicted and measured values Devore, [2005]. For the current problem, the following result is obtained: R = 0.999 in absence of acetic acid (presence of the protective film formation) as shown in Figure 4.15.

110

Chapter Four

Results & Discussion 40 38 36

C.R Predicted (gmd)

34 32 30 28 26 24 22 20 18

Experimental Data Points Quadratic Model (R = 0.999)

16 14 12 10 10

12

14

16

18

20

22

24

26

28

30

32

34

36

38

40

C.R Measured (gmd)

Fig. 4.15 Performance of the Quadratic Mathematical Model for API X65 Mild Steel Alloy in Absence of Acetic Acid (Presence of Protective Film Formation).

4.1.4.6 Graphic Analysis of the Model: The aim of this study was to find a corrosion rate whose features would have been previously defined from the operative conditions extracted from the quadratic mathematical model. Because the direct exploitation of the Equation was delicate, it was convenient to restore it under a graphic representation; while fixing two of the three factors of the survey, it was possible to represent the response surface materializing the surface of regression in a three-dimensional space. It was also possible to project the Equation in a design under isoresponse curves, interpreted as card curves level. (i) Evolution of Corrosion Rate as a Function of the Temperature and the pH: Figure 4.16 shows the synergism between the two factors: the temperature and pH in corrosion rate at speed of rotation 1425.8 rpm. It can be noted that the effect of the temperature differed according to the corrosion rate’s variation. This effect becomes positive and even more important when the corrosion rate is reduced. Considering simultaneous effects of temperature and pH is presented in Figure (4.16 contour plot). 111

Chapter Four


The Figure shows, in low pH (pH 7.5), the increase of corrosion rate is higher than in higher pH (pH 8.5).

60 50


40 30 20 10

8.6

8.4

76

8.2

pH

74

72

8.0 68

7.6

7.4

66

64

Te

o

e( tur e ra p m 70

7.8

C)

> < < < <
< < < < < < <
< < < < <
80 kJ/mol). At other conditions the results are between these values with intermediate values of activation energy, which indicate that the corrosion process is under mixed-control regimes. Perez, [2004] pointed out that if the change of the slow stage of the reaction was not taken into account, the Arrhenius Equation could not be applied. Robertson and Forrest, [2012] obtained values for activation energy for the corrosion protection of mild steel in deoxygenated acidic, neutral and alkaline aqueous chloride solutions. Ideally, a protective layer growth due to corrosion products deposition is a substance, which greatly increases the value of activation energy of corrosion process and formed on the metallic surface and prevents it to corroding. But, they found that some protective and non protective 126

Chapter Four


films did not affect the value of activation energy, therefore, it takes a constant mean value. Other protective layers reduce the value of activation energy. This suggests that the presence of these ironcarbonate protective-layers in corrosive medium modify the kinetics of corrosion reaction by offering alternate reaction paths with lower activation energies.

4.1.5.2 Effect of pH: Figure 4.26 shows the relationship between corrosion rates and pH at given temperatures at optimum conditions, in presence and absence of acetic acid. In presence of acetic acid, it is generally accepted to state: 1. The analysis of the response function showed that pH is slightly less pronounced than the temperature on the corrosion rate. 160

T = 40 o C T = 45 o C T = 50 o C T = 55 o C T = 60 o C T = 65 o C T = 67.5 o C T = 70 o C T = 72.5 o C T = 75 o C


140 120 100 80 60 40 20 0 2

3

4

5

6

7

8

9

pH

Fig. 4.26 The Effect of pH on the Corrosion of API X65 Mild Steel in CO2 saturated, 3.5 wt% NaCl Solution in Presence and Absence of Acetic Acid (Absence and Presence of the Protective Film Formation) at (2178.5 ppm HAc and 1296.6 rpm) and (1425.8 rpm). 2. The corrosion rate decreased as the pH value increased until it reaches pH 4.8 at 45 °C then slightly increased as the pH value increased. This behavior can be attributed to: When the pH value increased, the 127

Chapter Four


corrosion rate decreased because the formation of corrosion products scales (Fe3C+FeCO3). From Figure 4.26, it is clear that, increasing the pH from 3 to 4.8 at 45 °C gives an increase in the formation products scale layer greater than the increase in the dissolution of metal which increased the anodic reaction Uhlig and Winston, [2008] therefore the positive net from these two factors represented by the decrease in the corrosion rate. On the contrary, increasing the pH value up to pH 4.8 gave a negative net represented by the decrease in the formation of a very porous and nonprotective film thickness and slightly increases in the corrosion rate. Figure 4.26 shows also the corrosion rate in absence of acetic acid, presence of the protective film formation at the optimum conditions. It is generally accepted to state that: 1. The analysis of the response function showed that pH is less pronounced than the temperature on the corrosion rate, but showed that the interaction is the factor which has the lowest effect on the corrosion rate. 2. It can be seen that the corrosion rate slightly increases as the pH value increases until it reaches pH 7.9 at 68 °C then decreases as the pH value increases. This behavior can be attributed to: When the pH value increases, thereby increasing activity of H+ ions, indirectly accelerated corrosion kinetics Jiabin et al., [2011] and the corrosion rate increases therefore the dissolution rate of metal which increases the anodic reaction Uhlig and Winston, [2008]. On the other hand further increase in pH value results in the increase of the formation (deposition) rate of protective film thickness and covering (inhibiting) a portion of the steel surface due to protective film formation is accelerated by measures that restrict the transport of reaction products

128

Chapter Four


from the surface Daugstad et al., [2000] and decrease in the corrosion rate. Also, it is clear that, increasing the pH from 7.5 to 7.9 at 68 °C gave an increase in the dissolution of metal which increased the anodic reaction Uhlig and Winston, [2008] greater than the increase in the formation rate of corrosion product film therefore the positive net from these two factors represented by the increase in the corrosion rate. On the contrary, increasing the pH value up to pH 7.9 gave a negative net represented by the increase in the formation of a dense protective film thickness and decrease in the corrosion rate.

4.1.5.3 Effect of Acetic Acid Concentration: Figure 4.27 shows the relationship between corrosion rate and acetic acid concentration at different temperatures at optimum conditions. It can be stated that: (1) The analysis of the response function showed that concentration of acetic acid is less pronounced than the temperature on the corrosion rate, but showed that the interaction is the factor which has the lowest effect on the corrosion rate.

129

Chapter Four


75

70


65

60

55

50

T = 40 0C 45

T = 45 0C T = 50 0C

40

T = 55 0C T = 60 0C

35 800

1000

1200

1400

1600

1800

2000

2200

2400

2600

2800

3000

3200

Acetic Acid Concentration (ppm)

Fig. 4.27 The Effect of Acetic Acid Concentration on the Corrosion of API X65 Mild Steel in CO2 saturated, 3.5 wt% NaCl Solution in Presence of Acetic Acid (Absence of the Protective Film Formation) at (pH 4.8 and 1296.6 rpm). Also, it can be seen that the corrosion rate increased as the acetic acid concentration increased until it reaches 2178.5 ppm at 45 °C then decreased as the acetic acid concentration increased. This behavior can be attributed to: The significant increase of corrosion rate is due to the increase in cathodic reaction. As reported by Tran et al., [2013]. HAc increases cathodic reaction by dissociation of acetic acid and direct reduction of undissociated HAc molecules. Also, this is due to the degradation of non protective film by HAc, since acetate ions have the ability to form iron acetate and transport iron away from the steel surface Zhang and Cheng, [2009] then the corrosion rate decreased by non protective film formation (i.e., these films can significantly alter the corrosion process by either decreasing the corrosion rate by acting as a diffusion barrier, or increasing the corrosion Gulbrandsen et al., [1998] by increasing the active specimen surface area) is accelerated by

130

Chapter Four


measures that restrict the transport of reaction products from the surface Daugstad et al., [2000]. (2) Furthermore, an investigation shows that the corrosion rate rises at higher temperatures and HAc concentrations. This is due to acceleration of anodic and cathodic reactions when the temperature increases George, [2003] and George and Nesic, [2007]. The increase of cathodic reaction in CO2 corrosion is due to the HAc contribution to hydrogen ions through possibly dissociation and reduction. It is worthwhile to note that below the inhibitive level, the higher the concentration of acetic acid, the higher is the number of hydrogen ions produced. At higher temperature, the diffusion coefficient of HAc is higher, which results in more available species, approximately a twofold increase in the corrosion rate with 2000 ppm HAc at 60 °C than at 40 °C. At 40 °C, the value of the diffusion coefficient of HAc is 1.6x10 -9 m2/s, whereas at 60 °C it is 2.3x10-9 m2/s, which is 85% higher. (3) It is clear that, increasing the concentration of acetic acid from 1000 to 2178.5 ppm at 45 °C gave an increase in the corrosion rate of metal surface greater than the increase in the formation rate of corrosion scale film, therefore the positive net from these two factors represented by the increase in the corrosion rate. On the contrary, increasing the concentration of acetic acid up to 2178.5 ppm gave a negative net represented by the decrease in the corrosion rate by non protective film formation by acting as a diffusion barrier Gulbrandsen et al., [1998] and accelerated by measures that restrict the transport of reaction products from the surface Daugstad et al., [2000]. (4) A different corrosion rate trend is observed with the presence of HAc at range (1000-3000 ppm) under turbulent conditions at different 131

Chapter Four


temperatures. The corrosion rate increases with increasing HAc concentration. However, at 45 °C and pH 4.8, the corrosion rate decreases with more than 2178.5 ppm HAc. Furthermore, it is worth noting that an appreciable effect of HAc at 45 °C and pH 4.8 was only observed beyond 2178.5 ppm HAc, which recorded an approximately 12% increase in the corrosion rate. Figure 4.27 shows that increase of HAc concentration and temperature leads to an increase of corrosion rate. In the range of experiments, at optimum temperature 45 °C have increased corrosion rate to 48.9 gmd. This increase of corrosion rate as an effect of HAc concentration and temperature was also observed by Mokhtar, [2005] and James, [2004] and George and Nesic, [2004]. They all agreed with the role of ions activities that took part in contributing on corrosion rate.

4.1.5.4 Effect of Speed of Rotation: Figure 4.28 shows the effect of speed of rotation on corrosion rate of mild steel in absence and presence of the protective film at different temperatures at optimum conditions. In absence of the protective film, it is clear that: (1) The speed of rotation is less pronounced than the temperature on the corrosion rate, but showed that the interaction is the factor which has the lowest effect on the corrosion rate.

132

Chapter Four


80

T = 40 oC T = 45 oC


70

T = 50 oC T = 55 oC

60

T = 60 oC T = 65 oC

50

T = 67.5 oC T = 70 oC

40

T = 72.5 oC T = 75 oC

30

20

10 900

1000

1100

1200

1300

1400

1500

1600


Fig. 4.28 The Effect of Speed of Rotation on the Corrosion of API X65 Mild Steel in CO2 saturated, 3.5 wt% NaCl Solution in Presence and Absence of Acetic Acid (Absence and Presence of the Protective Film Formation) at (pH 4.8 and 2178.5 ppm HAc) and (pH 7.9). (2) It can be seen that the corrosion rate increased as the speed of rotation increased until it reaches 1296 rpm at 45 °C then decreased as the speed of rotation increased. This behavior can be attributed to: When the speed of rotation increased, the corrosion rate increased and resulted in an enhanced value of the corrosion rate on pure kinetic ground because of the speed of rotation lead to removal any layers, corrosion products or deposits from the surface and get a more active surface (i.e., the formation rate of corrosion products scale (Fe3C+FeCO3) non protective film decreased). On the other hand further increase in speed of rotation results in the increase of the formation rate of corrosion products on the metal surface and decrease corrosion rate because for a not fully developed protective film surface, the effect of speed of rotation is related to the transport of species towards and away from the metal surface. (3) It is clear that, increasing the speed of rotation from 1000 to 1296 rpm at 45 °C gave an increase in the dissolution of metal which increases 133

Chapter Four


the anodic reaction Uhlig and Winston, [2008] greater than the increase in the formation rate of corrosion scale therefore the positive net from these two factors represented by the increase in the corrosion rate. On the contrary, increasing the speed of rotation up to 1296 rpm gave a negative net represented by the decrease in the corrosion rate. From Figure 4.28, in presence of the protective film formation. It is clear that: (1) The speed of rotation is less pronounced than the temperature on the corrosion rate, but showed that the interaction is the factor which has the highest effect on the corrosion rate. (2) It can be seen that the corrosion rate increased as the speed of rotation increased until it reaches 1425 rpm at 68 °C then slightly decreased as the speed of rotation increased. This behavior can be attributed to: When the speed of rotation increased, the corrosion rate increased because the formation rate of corrosion products scale (FeCO3) protective film thickness decreased. On the other hand further increase in speed of rotation results in the increase of the formation rate of protective film thickness and decrease corrosion rate. (3) It is clear that, increasing the speed of rotation from 1000 to 1425 rpm at 68 °C gave an increase in the dissolution of metal which increased the anodic reaction Uhlig and Winston, [2008] greater than the increase in the formation rate of corrosion scale therefore the positive net from these two factors represented by the increase in the corrosion rate. On the contrary, increasing the speed of rotation up to 1425 rpm gave a negative net represented by the decrease in the corrosion rate by protective film formation is accelerated by measures that restrict the transport of reaction products from the surface Daugstad et al., [2000].

134

Chapter Four


(4) It can be seen that the protective film thickness decreased and corrosion rate increased as the speed of rotation increased until it reaches 1425 rpm which is consistent with Faraday’s law then film thickness increased as the speed of rotation increased. This behavior can be attributed to: When the speed of rotation increased, the dissolution of metal which increased the anodic reaction Uhlig and Winston, [2008], because the protective film formation process on steel surface include three essential stages where these need to accelerate corrosion, during these stages can show speed of rotation influence on protective film thickness, essential stages for protective layer formation include Nesic , [2007]: a. Attacked stage of corrosive solution to the steel surface and accumulation of layers of insoluble corrosion products on the steel surface. b. Formation stage of a diffusion barrier for the species involved in corrosion process. c. Formation stage of protective layer (dense layer). At speed of rotation 1425 rpm the dissolution rate of metal surface is equal to the formation rate of the protective film. Thus the mathematical optimum speed of rotation is equal to 1425 rpm.

4.1.6 Combined Influence of Temperature, pH, Acetic Acid Concentration and Speed of Rotation on the Corrosion Rate: The experimental corrosion rate results in presence and absence of acetic acid for CO2 saturated, 3.5 wt% NaCl solutions as a function of temperature, pH, HAc acid concentration and speed of rotation are given in Tables 4.2 and 4.7 respectively. Many relationships can be suggested to correlate these variables.

135

Chapter Four


Second order polynomial model was used to represent these variables as follows:

4.1.6.1 No Protective Film Formation:

(

)

Where;

: Corrosion Rate (gmd) : Temperature (°C) : pH : Acetic Acid Concentration (ppm) : Speed of Rotation (rpm) And, b0, b1, b2, b3…b14: are constants.

Using non-linear estimation regression by means of the least-square method, using STATISTICA program software package® version 10 to evaluate these constants. Equation (4.17) used as a first model (L1) in no protective film formation (in presence of acetic acid) to represent the results. The corrosion rate results can be related to temperature by Arrhenius

type

Equation,

which

expressed

as

 E  Corrosion Rate C.R.  A Exp   a   RT

follows: (2.36)

It is known also, that the corrosion rate decreases as the temperature decreased, as pH increased, as HAc acid concentration decreased and as speed of rotation decreased, so; (

(

)) (

) (

) (ω)

(

)

Where T is temperature in (°C), pH is hydrogen ion concentration, CA is HAc acid concentration in (ppm) and ω is speed of rotation in (rpm). So, it is suggested a new model to correlate the corrosion rate data (gmd), as a function of temperature (K), pH, HAc acid concentration 136

Chapter Four


(ppm) and speed of rotation (rpm), through the following proposed Equation; (

(

)) (

) (

) (ω)

(

)

Then, the second model (L2) suggested in term of Y, X1, X2, X3 and X4 as in the first model is; (

(

)) (

(

)

)

Where; b15, b16, b17, b18 and b19 are other constants. The non-linear regression for L1 and L2 yields the following Equations:The First Model, L1, is:

(

)

Correlation Coefficient, R = 0.999 And the Second Model, L2, is: (

(

)) (

)

(

)

Correlation Coefficient, R = 0.965 The constant b16, in Equation (4.20), is the Arrhenius slope (-Ea /R), in Equation (4.20a) this value is equal to (-3793.119), then the activation energy can be obtained, as Ea = 31.536 kJ/mole, while the average value of activation energy obtained at different conditions in presence of acetic acid and shown in Table 4.12, (i.e., Ea ranged from (16.37 - 41.01 kJ/mol), with Eav = 30.181 kJ/mole). The two models can represent the data in absence of protective film formation, with more accuracy of L1 than L2, as shown in Figure 4.29.

137

Chapter Four

Results & Discussion In In In In

140

L1 In Absence of Protective Film L2 In Absence of Protective Film L3 In Presence of Protective Film L4 In Presence of Protective Film

120

Calculated (gmd)

Absence of Protective 1: Y = 0.0805 + 0.9988 * X; R = 0.999 pred.Film M1 = L0.0805+0.9988*x pred.Film M2 = L0.1116+0.9873*x Absence of Protective 2: Y = 0.1116 + 0.9873 * X; R = 0.967 pred. Film M1 = 0.0015+0.9999*x Presence of Protective L3: Y = 0.0015 + 0.9999 * X; R = 1.000 pred. M2 = 1.639+0.9261*x Presence of Protective Film L4: Y = 1.639 + 0.9261 * X; R = 0.948

100 80 60 40 20 0 0

20

40

60

80

100

120

140

Experimental (gmd)

Fig. 4.29 The Deviation of the Two Models Suggested from the Corrosion Rate Data of API X65 Mild Steel in CO2 Saturated, 3.5 wt% NaCl Solution in Absence and Presence of Protective Film Formation.

4.1.6.2 In Presence of Protective Film: (

)

Where;

: Corrosion Rate (gmd) : Temperature (°C) : pH : Speed of Rotation (rpm) And, b0, b1, b2, b3…b9: are constants.

Using non-linear estimation regression by means of the least-square method, using STATISTICA program software package® version 10 to evaluate these constants. Equation (4.21) used as a first model (L3) in presence of protective film (in absence of acetic acid) to represent the results. The corrosion rate results can be related to temperature by Arrhenius

type

Equation,

which

 E  Corrosion Rate C.R.  A Exp   a   RT

138

expressed

as

follows: (2.36)

Chapter Four


It is known also, that the corrosion rate decreases as the temperature decreased, as pH increased and as speed of rotation decreased, so; (

(

)) (

) ( )

(

)

Where T is temperature in (°C), pH is hydrogen ion concentration and ω is speed of rotation in (rpm). So, it is suggested a new model to correlate the corrosion rate data (gmd), as a function of temperature (K), pH and speed of rotation (rpm), through the following proposed Equation; (

(

)) (

) (ω)

(

)

Then, the second model (L4) suggested in term of Y, X1, X2 and X4 as in the first model is; (

(

)) (

(

)

)

Where; b10, b11, b12 and b13 are other constants. The non-linear regression for L3 and L4 yields the following Equations:The First Model, L3, is:

(

)

(

)

Correlation Coefficient, R = 0.999 And the Second Model, L4, is:(

(

)) (

)

Correlation Coefficient, R = 0.915 The constant b11, in Equation (4.24), is the Arrhenius slope (-Ea /R), in Equation (4.24a) this value is equal to (-7239.757), then the activation energy can be obtained, as Ea = 60.1913 kJ/mole, which is deviate from 139

Chapter Four


the average value of activation energy obtained at different conditions in absence of acetic acid and shown in Table 4.12, (i.e. Ea ranged from (49.44 - 115.19 kJ/mol), with Eav = 72.49 kJ/mole). These suggest that, L4 represent the data in presence of protective film with low correlation coefficient, this may be due to high differences between the corrosion rates in absence and presence of protective film. L3 can be used to represent the corrosion rate data in presence of protective film formation, as shown in Figure 4.29. It is clear, that the value of activation energy obtained from Equation (4.20) is approach the average value obtained from Table 4.12 at different conditions in presence of acetic acid. While, in absence of acetic acid, the value of activation energy obtained from Equation (4.24) is slightly deviate from the average value obtained from Table 4.12 at different conditions. Or in other words, when there is no effect of acetic acid concentration on the corrosion rate. So, Equation (4.24), gives low correlation coefficient in Arrhenius Equation, and L3 is a more accurate than L4.

4.2 Electrochemical Results: 4.2.1 Open Circuit Potential Measurements (OCP): The values of OCP were measured from the experimental runs. The steady state potentials were reached within one hour. Table 4.13 summarizes the values of OCP at different conditions. The measure OCP are found very close to the values given in the literature Zhang and Cheng, [2009] and Tran et al., [2013].

140

Chapter Four


Tables 4.13 Open Circuit Potential Values in Presence and Absence of Acetic Acid at Different Conditions. Run No.

Temp. (°C)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

40 50 60 40 50 60 40 50 60 40 50 60 40 50 60 40 50 60 65 70 75 65 70 75 65 70 75 65 70 75 65 70 75 65 70 75

pH

HAc Conc. (ppm)


3

1000

1000

3

1000

1250

3

1000

1500

5

3000

1000

5

3000

1250

5

3000

1500

7.5

Blank Solution

1000

7.5

Blank Solution

1250

7.5

Blank Solution

1500

8.5

Blank Solution

1000

8.5

Blank Solution

1250

8.5

Blank Solution

1500

141

OCP (mV),SCE -560 -565 -585 -573 -582 -600 -597 -604 -613 -523 -529 -548 -529 -548 -567 -559 -560 -568 -412 -435.5 -465 -414 -439 -471 -422 -447.5 -481 -345 -398 -408 -406 -429 -444 -413 -440 -446

Chapter Four


It is clear that: 1. In presence of 1000 ppm & 3000 ppm acetic acid at pH 3 & pH 5 respectively, the OCP potential shifts to more negative (more active) direction as the temperature increased from 40 to 60 °C associated with speed of rotation increases from 1000 to 1500 rpm. 2. The same behavior observed in absence of acetic acid at pH’s (7.5 and 8.5), the increasing in temperature and speed of rotation lead to shift the OCP potential to more negative (active) values. But the OCP in absence of HAc are more positive in comparison with in presence of acetic acid. This shift towards higher anodic potential can be explained by the possible formation of an iron carbonate film and by the effect of temperature, pH and speed of rotation on the cathodic diffusion current. There is occasionally a change in Tafel slopes of polarization curve, more usually the cathodic one, possibly because of alteration of structure of double layer attendant upon protective film formation and hence changes in α (i.e. symmetry coefficient) Bagotsky, [2006]. 3. The recorded open circuit potential values were close to those measured by Zhang and Cheng, [2009] and Tran et al., [2013] at nearly similar test conditions.

4.2.2 API X65 Mild Steel Potentiodynamic Polarization Curves: The corrosion behavior of API X65 mild steel in CO2 saturated, NaCl solution in presence and absence of acetic acid was studied using polarization measurements. A total of 36 runs were carried out at different conditions as shown in Table 4.14.

142

Chapter Four


Table 4.14 Conditions of Potentiodynamic Polarization in Presence and Absence of Acetic Acid at Different Conditions. Run No.

Temp. (°C)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

40 50 60 40 50 60 40 50 60 40 50 60 40 50 60 40 50 60 65 70 75 65 70 75 65 70 75 65 70 75 65 70 75 65 70 75

pH

HAc Conc. (ppm)


3

1000

1000

3

1000

1250

3

1000

1500

5

3000

1000

5

3000

1250

5

3000

1500

7.5

Blank Solution

1000

7.5

Blank Solution

1250

7.5

Blank Solution

1500

8.5

Blank Solution

1000

8.5

Blank Solution

1250

8.5

Blank Solution

1500

143

Chapter Four


4.2.2.1 API X65 Potentiodynamic Polarization Curves (No Protective Film Formation): Figures 4.30 through 4.41 showed the polarization curves obtained for API X65 in CO2 saturated, NaCl solution, at different temperatures and speeds of rotation. A similar behavior is observed for the whole range of temperatures at all speeds of rotation and acetic acid concentrations analyzed. A plateau corresponding to a limiting current density (ilim) is observed in the cathodic region of the polarization curves between (1000, -700 mV SCE) at pH = 3. Therefore the cathodic reaction seems to be controlled by diffusion. Similar behavior can be seen at pH = 5, but the plateau corresponding to (ilim.) current density is not so clear compared with pH = 3. With regard to the anodic branch of the polarization curves, anodic current density continuously increases as potential shifts to more positive values, not reaching stability. This fact means that the API X65 mild steel experiencing a corrosion process and it cannot reach a passivation state. -100 -200

Potential (mV) vs. SCE

-300 -400 -500 -600 -700

40 o C 50 o C 60 o C

-800 -900 -1000 -1100 -2

-1

0

1

2

3

4

5

Log i ( µA/cm2 )

Fig. 4.30 Polarization Curves for the Corrosion of API X65 Mild Steel in CO2 saturated,3.5 wt % NaCl in Presence of Acetic Acid at pH 3, 1000 ppm HAc and Speed of Rotation 1000 rpm.

144

Chapter Four

Results & Discussion -200 -300


-400 -500 -600 -700 -800 o

40 C -900

o

50 C o

60 C

-1000 -1100 -2

-1

0

1

2

3

4

5

Log i ( µA/ cm2 )


-200 -300


-400 -500 -600 -700 -800

40 o C 50 o C 60 o C

-900 -1000 -1100 -2

-1

0

1

2

3

4

5

2

Log i ( µA/cm )


145

Chapter Four

Results & Discussion -100 -200


-300 -400 -500 -600 -700 -800

1000 rpm 1250 rpm 1500 rpm

-900 -1000 -1100 -2

-1

0

1

2

3

4

5

Log i ( µA/cm2 )

Fig. 4.33 Polarization Curves for the Corrosion of API X65 Mild Steel in CO2 saturated, 3.5 wt % NaCl in Presence of Acetic Acid at pH 3, 1000 ppm HAc and Temperature 40°C.

-200 -300

Potential ( mV ) vs. SCE

-400 -500 -600 -700 -800

1000 rpm 1250 rpm 1500 rpm

-900 -1000 -1100 -2

-1

0

1

2

3

4

5

2

Log i ( µA/cm )


146

Chapter Four

Results & Discussion -300


-400

-500

-600

-700

-800

1000 rpm 1250 rpm 1500 rpm

-900

-1000

-1100 -2

-1

0

1

2

3

4

5

Log i ( µA/cm2 )


0


-200

-400

-600

-800

T = 40 o C T = 50 o C T = 60 o C

-1000

-1200 -2

-1

0

1

2

3

4

5

2

Log i ( µA/cm )

Fig. 4.36 Polarization Curves for the Corrosion of API X65 Mild Steel in CO2 saturated, 3.5 wt % NaCl in Presence of Acetic Acid at pH 5, 3000 ppm HAc and Speed of Rotation 1000 rpm.

147

Chapter Four


0 -100


-200 -300 -400 -500 -600 -700

T = 40 o C T = 50 o C T = 60 o C

-800 -900 -1000 -1100 -2

-1

0

1

2

3

4

5

2

Log i ( µA/cm )


-100 -200


-300 -400 -500 -600 -700 -800

T = 40 o C T = 50 o C T = 60 o C

-900 -1000 -1100 -2

-1

0

1

2

3

4

5

2

Log i ( µA/cm )


148

Chapter Four


0


-200

-400

-600

-800

1000 rpm 1250 rpm 1500 rpm

-1000

-1200 -2

-1

0

1

2

3

4

5

2

Log i ( µA/cm )


0


-200

-400

-600

-800

1000 rpm 1250 rpm 1500 rpm

-1000

-1200 -2

-1

0

1

2

3

4

5

2

Log i ( µA/cm )

Fig. 4.40 Polarization Curves for the Corrosion of API X65 Mild Steel in CO2 saturated, 3.5 wt % NaCl in Presence of Acetic Acid at pH 5, 3000 ppm HAc and Temperature 50°C. 149

Chapter Four


0 -100


-200 -300 -400 -500 -600 -700 -800

1000 rpm 1250 rpm 1500 rpm

-900 -1000 -1100 -2

-1

0

1

2

3

4

5

Log i ( µA/cm2 )


4.2.2.2 API X65 Potentiodynamic Polarization Curves (Under Protective Film Formation): Figures 4.42 through 4.53 show the polarization curves in presence of protective film formation, at different temperatures, solution pH values and speeds of rotation. In this case the cathodic region of the polarization curves, and the cathodic reactions occurred seems to be activation complicated by diffusion (mass transfer), leading to disappearance of the plateau corresponding to a limiting current density. With regards to the anodic branch of the polarization curves. Similar behavior as in absence of protective film can be seen clearly. (i.e., the higher the potential shift to the positive direction, the higher the anodic current density leading to no passivity state reaching).

150

Chapter Four


0


-200

-400

-600

-800

T = 65 o C T = 70 o C T = 75 o C

-1000

-1200 -1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

2

Log i (µA/cm )

Fig. 4.42 Polarization Curves for the Corrosion of API X65 Mild Steel in CO2 saturated, 3.5 wt % NaCl in Absence of Acetic Acid at pH 7.5 and Speed of Rotation 1000 rpm.

0


-200

-400

-600

-800

T = 65 o C T = 70 o C T = 75 o C

-1000

-1200 -1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

2

Log i ( µA/cm )


151

Chapter Four


0


-200

-400

-600

-800 o

T = 65 C o

T = 70 C -1000

-1200 -1.5

o

T = 75 C

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Log i ( µA/cm2 )


0


-200

-400

-600

-800

1000 rpm 1250 rpm 1500 rpm

-1000

-1200 -1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Log i ( µA/cm2 )

Fig. 4.45 Polarization Curves for the Corrosion of API X65 Mild Steel in CO2 saturated, 3.5 wt % NaCl in Absence of Acetic Acid at pH 7.5 and Temperature 65°C.

152

Chapter Four


0


-200

-400

-600

-800

1000 rpm 1250 rpm 1500 rpm

-1000

-1200 -1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

2

Log i ( µA/cm )


0


-200

-400

-600

-800

1000 rpm 1250 rpm 1500 rpm

-1000

-1200 -1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Log i ( µA/cm2 )

Fig. 4.47 Polarization Curves for the Corrosion of API X65 Mild Steel in CO2 saturated, 3.5 wt % NaCl in Absence of Acetic Acid at pH 7.5 and Temperature 75°C. 153

Chapter Four


0


-200

-400

-600

-800

T = 65 o C T = 70 o C T = 75 o C

-1000

-1200 -1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Log i ( µA/cm2 )


0


-200

-400

-600

-800 o

T = 65 C o

T = 70 C -1000

-1200 -1.5

o

T = 75 C

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Log i ( µA/cm2 )

Fig. 4.49 Polarization Curves for the Corrosion of API X65 Mild Steel in CO2 saturated, 3.5 wt % NaCl in Absence of Acetic Acid at pH 8.5 and Speed of Rotation 1250 rpm. 154

Chapter Four



-200

-400

-600

-800

T = 65 o C T = 70 o C T = 75 o C

-1000

-1200 -1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

2

Log i ( µA/cm )


0


-200

-400

-600

-800

1000 rpm 1250 rpm 1500 rpm

-1000

-1200 -1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

Log i ( µA/cm2 )


155

Chapter Four


0


-200

-400

-600

-800

1000 rpm 1250 rpm 1500 rpm

-1000

-1200 -1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

2

Log i ( µA/cm )


0


-200

-400

-600

-800

1000 rpm 1250 rpm 1500 rpm

-1000

-1200 -1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Log i ( µA/cm2 )

Fig. 4.53 Polarization Curves for the Corrosion of API X65 Mild Steel in CO2 saturated, 3.5 wt % NaCl in Absence of Acetic Acid at pH 8.5 and Temperature 75°C. 156

Chapter Four


4.2.3 Parameters Estimated from API X65 Polarization Curves: 4.2.3.1 Corrosion Potentials and Corrosion Current Densities (Tafel Extrapolation Method): Figures 4.30 through 4.53 show the polarization curves in absence and presence of the protective film formation at different experimental conditions. The corrosion current densities and corrosion potentials were estimated by Tafel extrapolation of cathodic and anodic curves to the corrosion potentials. The Tafel slopes were estimated as well. Table 4.15 shows the corrosion current densities, corrosion potentials and cathodic and anodic Tafel slopes at different experimental conditions. In all cases the extrapolation started over about 50 mV away from Ecorr, and the same range of potential values has been always used. Therefore it has been considered that the obtained icorr values are accurate enough to study the influence of temperature, pH, acetic acid concentration and speed of rotation on corrosion process. The characteristic of cathodic polarization curves with variation of temperature and speed of rotation have been illustrated in mentioned Figures 4.30 through 4.53. These polarization curves provide information about effects of changes in potential on the corrosion of the cathode as current density (current per unit area). Since the electrolyte is seawater (salt water), the concentration polarization type is predominant Jezmar, [2002]. From polarization curves, it can be determined practically the free corrosion potential, Ecorr and limiting current density, ilim.. Where Ecorr is determined when the potential becomes approximately constant with decreasing current. The limiting current plateau is not well defined, thus the method given by Gabe and Makanjoula, [1986] will be adopted to find ilim.values: ( 157

)

Chapter Four


where, i1 and i2 are the current associated with E1 and E2 respectively as shown in Figure 4.54.

Fig. 4.54 the Limiting Current Density by Gabe and Makanjoula Method Gabe and Makanjoula, [1986].

158

Chapter Four


Table 4.15 Corrosion Parameters Obtained (by Tafel Extrapolation Method) from Polarization Curves at Different Conditions. Run No.

Temp. (°C)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

40 50 60 40 50 60 40 50 60 40 50 60 40 50 60 40 50 60 65 70 75 65 70 75 65 70 75 65 70 75 65 70 75 65 70 75

pH

HAc Conc. (ppm)


3

1000

1000

3

1000

1250

3

1000

1500

5

3000

1000

5

3000

1250

5

3000

1500

7.5

Blank Solution

1000

7.5

Blank Solution

1250

7.5

Blank Solution

1500

8.5

Blank Solution

1000

8.5

Blank Solution

1250

8.5

Blank Solution

1500

ba (mV)

-bc (mV)

icorr (µA/cm2)

Ecorr (mV,SCE)

ilim (µA/cm2)

123.9 141.1 154.6 136.9 148.8 152.8 137.5 148.9 150.5 138.5 148.6 159.4 135.9 148.7 152.0 136.9 148.8 151.0 91.8 82.7 73.4 91.3 82.4 73.2 91.5 79.3 67.2 95.3 85.1 73.3 90.9 82.3 73.0 96.4 84.5 76.2

347.3 421.4 575.3 348.4 435 550 349 428 595 334 425 580 338 475 598 343 433 578 285 393 415 227 325 444 237 308 424 221 320 439 224 338 464 226 328 468

200.96 255.76 495.43 233.91 292.58 552.56 257.49 354.36 570.28 176.97 185.51 214.99 177.14 197.61 233.36 190.95 215.03 231.01 70.30 98.03 132.3 76.50 110.57 134.44 88.92 123.03 138.71 56.51 96.27 117.60 64.34 101.13 119.48 72.17 105.98 121.37

-572 -580 -600 -582 -596 -616 -610 -620 -628 -536 -544 -563 -545 -562 -576 -570 -575 -580 -423 -451.5 -480 -427 -456 -485 -430 -462.5 -495 -353 -412 -425 -414 -445 -455 -425 -455 -460

251.77 290.38 601.91 563.74 1124.79 2324.18 706.37 1365.88 2391.48 485.05 538.84 787.12 437.63 565.66 1807.69 549.59 823.92 2271.21 186.96 224.52 282.38 204.59 235.15 256.48 172.53 224.67 297.72 163.58 226.52 297.18 172.01 180.29 245.46 171.06 283.38 319.73

159

Chapter Four


Table 4.15, generally, shows that: 1. The values of corrosion current density increase with increasing in temperatures and speeds of rotation in both the presence and absence of acetic acid (i.e., absence and presence of the protective film formation) at different experimental conditions. But the corrosion current densities in absence of HAc are small in comparison with in presence of acetic acid. Temperature increases the rate of almost all chemical

reactions

Shrier

et

al.

(2),

[2000]

and

Uhlig,

[2011].Therefore the effect of temperatures is to increase the anodic and cathodic current in presence and absence of acetic acid for a given pH value, HAc acid concentration and speed of rotation as shown by Figures 4.30 through 4.32, 4.36 through 4.38, 4.42 through 4.44 and 4.48 through 4.50. 2. The values of Ecorr shifts to more negative (more active) direction with increasing in temperatures and speeds of rotation in both the presence and absence of acetic acid at different experimental conditions. But the corrosion potentials in absence of HAc are more positive in comparison with in presence of acetic acid. The corrosion potential was shifted to more negative values (more active) with increasing temperature in presence and absence of acetic acid for a given pH value, HAc acid concentration and speed of rotation. The anodic and cathodic currents have increased with temperature as shown schematically in Figure 4.55.

160

Chapter Four


Fig. 4.55 Effect of Temperature on Corrosion Potential and Corrosion Current Andijani and Turgoose, [1999]. Nervana, [2010] pointed out that the main effect of increasing the temperature is to increase the exchange current, when the process is under activation control. George and Nesic, [2007] and Hernandez et al., [2012] presented the analysis of activation controlled cathodic reaction complicated by mass transport effect and discussed its influence on corrosion parameters. 3. In both presence and absence of acetic acid, the values of limiting diffusion current density were increased with both temperature and speed of rotation increasing. But the limiting diffusion current densities in absence of HAc are smaller in comparison with in presence of acetic acid. 4. The calculated values of the anodic Tafel slopes, ba in presence and absence of acetic acid at different experimental conditions, are shown as a function of the temperature, pH, HAc acid concentration and speed of rotation of the electrode. As can be observed, in general terms, as the speed of rotation increases, the measured ba remains relatively constant. This provides experimental evidence on the 161

Chapter Four


charge-transfer control of the anodic reaction, and therefore a mechanism that is not affected by speed of rotation. 5. The values of the calculated cathodic Tafel slopes bc, in this region are dependent on the data range considered. However, they are found to fall between -221 to -598 mV shown in Table 4.15. The variance in cathodic Tafel slopes in absence and presence of the protective film formation may be ascribed to changes in the symmetry of the energy barrier ( departs from 0.5) and the temperature Bagotsky, [2006]. 6. In general, in presence of the protective film formation at pH 8.5 and different speeds of rotation (1000, 1250 and 1500 rpm), the values of bc and ba were approximately unchanged compared with the values at pH 7.5 for the same conditions. However, in the experiments presented here, the measured Tafel slopes were very different from the expected 30 to 40 mV. The measured values are always higher than 120 mV. These values suggest that the metal surface could be covered by a film. Anodic dissolution of iron followed Tafel behavior for small overpotentials (slope 50-60 mV) and was not sensitive to speed of rotation. This would interfere with the anodic reaction by n hindering the escape of metal ions M  from the metal surface to the

solution, and the attack on the metal would then proceed at a reduced rate from less active anodic sites. 7. The effect of the formation of the protective film on anodic Tafel slopes, ba, is best understood by considering a potential energy diagram for (M+n) ions leaving the metal surface and approaching the hydrate state in solution. This is illustrated in Figure 4.56 which shows the relationship between the curves for (M+n) ions from very active and less active sites, with symmetry factors, α1 and α2. It is evident that as the more active sites are blocked by formed film ions and corrosion proceeds at less active site, the average symmetry 162

Chapter Four


2.303RT   ba  F factor,  , would increase. Since 

   , then, ba would reduce.

The positively charged layer of protective film ions would be an additional factor increasing  , and reducing ba, since it would n lengthen the path to be traveled by the escaping M  ions in an

Free Energy of M+n Ions

adverse field before reaching the aqueous layer.

Fig. 4.56 Potential Energy Diagram for Transition of M+n Ions from Metal Surface to Aqueous layer Driver and Meakins, [1974]. From Very Active Sites From Less Active Sites Hernandez et al., [2012] reported ba values in the range 149-162 mV for the corrosion of API X70 mild steel in deaerated 3% NaCl brine in absence of acetic acid at 20 °C, pH 4 and different speeds of rotation ranged (100-5000 rpm). Tafel slope of the hydrogen evolution reaction is a function of pH value and temperature and it is not affected by speed of rotation conditions as it is a charge transfer property Bala, [1988]. Most of CO2 corrosion reaction protective products and the coverage of the metal surface with the resulted scales caused ba to decrease, further increasing in temperature and pH value may lead to further decreasing of ba to a minimum values, then may lead to increase the value of b a again. 163

Chapter Four


This indicates that the formation of protective film plays an important part in causing the decrease ba Hernandez et al., [2012]. Also the changes in Tafel slopes (bc & ba) may be due either to removing of the protective film and / or to change in kinetic of dissolution of metal Sun et al., [2003].

164

Chapter Four


4.2.3.2 Corrosion Potentials and Corrosion Current Densities (McLaughlin Method): Figures 4.30 through 4.53 show the polarization curves in absence and presence of the protective film formation at different experimental conditions. In order to estimate the values of corrosion current density and Tafel slopes, potential against current density data can be used as input and output of Equation (2.73). Nonlinear regression of Equation (2.73) was used in order to obtain the parameters of this Equation. The regression method based on Rosenbrock and Quasi-Newton estimation method was used. The results of the method were shown in Table 4.16. Table 4.16 gives the average values of icorr, ba and bc corresponding to various

values

of

over

potential

(η)

interval

(i.e.,

at

10,20,30,40,50,60,70,80,90,100 mV SCE) from Ecorr of corrosion kinetic parameters were shown in Table 4.16. The current-potential curve of corroding metal is rather complex non-linear Equation; hence a general analytical solution is equally complex. For application of Equation (2.73), values of

and

were calculated by using the method given by

Gabe and Makanjoula, [1986] as described in Figure 4.54 for cathodic and anodic limiting current density since there is indication of concentration polarization for both with cathodic and anodic reactions. The results are more accurate with polarization curves at pH’s (3 and 5) under no protective film formation in Figures 4.30 through 4.41; the corrosion parameters are closer than data of Figures 4.30 through 4.41 compared with Figures 4.42 through 4.53 that less accurate. This may be attributed to that Equation (2.73) applicable with systems contains concentration polarization more than the system with activation complicated by diffusion (mass transfer) or the system with activation process only Khadom, [2010]. 165

Chapter Four


Table 4.16 Corrosion Parameters Obtained (by McLaughlin Method) from Polarization Curves at Different Conditions. Run No.

Temp. (°C)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

40 50 60 40 50 60 40 50 60 40 50 60 40 50 60 40 50 60 65 70 75 65 70 75 65 70 75 65 70 75 65 70 75 65 70 75

pH

HAc Conc. (ppm)


3

1000

1000

3

1000

1250

3

1000

1500

5

3000

1000

5

3000

1250

5

3000

1500

7.5

Blank Solution

1000

7.5

Blank Solution

1250

7.5

Blank Solution

1500

8.5

Blank Solution

1000

8.5

Blank Solution

1250

8.5

Blank Solution

1500

166

ba (mV)

-bc (mV)

icorr (µA/cm2)

85.8 95.1 102.6 91.9 98.8 101.2 92.6 98.9 100.5 93.5 98.5 104.4 90.9 98.5 102 91.6 98.5 100.8 71.4 66.5 64.1 69.8 67.5 64.2 69.9 63.9 60.2 71.9 67.5 63.1 69.9 66.8 60.9 70.8 70.3 69.4

373.7 505.7 632.6 376.7 505 619.5 382 512.7 647 376.5 502.5 637.5 370 536 645 373.1 511.2 634 358.1 505.1 556.3 325 473 590 334 455 573 318 465 585 324 475 609 324 475 602

197.6 251.9 490.5 230.5 288.7 557.5 254.2 350.5 575.1 173.5 181.5 219.8 173.6 194.1 238.3 187.3 211.2 235.6 66.5 93.3 135.1 72.7 104.3 138.7 83.8 120.4 145.1 53.2 90.9 120.8 60.1 96.5 125.1 86.1 100.7 125.5

Chapter Four


4.2.3.3 Corrosion Process Kinetic Parameters (β-Model): Corrosion process kinetic parameters determined from polarization curves by mathematical β-Model Korobove and Medvedeva, [2000] are listed in Tables 4.17 through 4.20. Table 4.17 Corrosion Process Kinetic Parameters as a Function of Temperatures and Speeds of Rotation obtained from β-Model in Presence of Acetic Acid (1000 ppm) at pH = 3. Run Temp. pH No. (°C)

1 2 3 4 5 6 7 8 9

HAc Speed of ba Conc. Rotation (mV) (ppm) (rpm)

40

3

1000

50

3

1000

60

3

1000

1000 1250 1500 1000 1250 1500 1000 1250 1500

47.6 46.9 47.5 49.1 48.8 48.9 50.6 49.8 50.5

-bc icorr (mV) (µA/cm2)

ilim. (µA/cm2)

β (icorr/ ilim.)

400 405 415 590 575 597 690 689 699

238.37 550.30 692.95 275.02 1109.43 1350.52 582.35 2304.62 2371.92

0.81 0.41 0.36 0.90 0.26 0.26 0.83 0.24 0.24

194.26 227.19 250.78 248.08 284.90 346.68 485.65 562.34 580.06

Table 4.18 Corrosion Process Kinetic Parameters as a Function of Temperatures and Speeds of Rotation obtained from β-Model in Presence of Acetic Acid (3000 ppm) at pH = 5. Run Temp. pH No. (°C)

1 2 3 4 5 6 7 8 9

HAc Conc. (ppm)

40

5

3000

50

5

3000

60

5

3000


1000 1250 1500 1000 1250 1500 1000 1250 1500

ba -bc icorr ilim. (mV) (mV) (µA/cm2) (µA/cm2)

48.5 45.9 46.9 48.6 48.7 48.8 49.4 52.0 51.0

167

419 402 404 580 597 590 695 698 694

170.26 170.43 184.24 177.83 189.93 207.35 224.77 243.14 240.79

471.63 424.21 536.17 523.48 550.30 808.54 767.54 1788.13 2251.65

β (icorr/ ilim.)

0.36 0.40 0.34 0.34 0.35 0.26 0.29 0.14 0.11

Chapter Four


Table 4.19 Corrosion Process Kinetic Parameters as a Function of Temperatures and Speeds of Rotation obtained from β-Model in Absence of Acetic Acid at pH=7.5. Run No.

1 2 3 4 5 6 7 8 9

Temp. (°C)

pH

65

7.5

70

7.5

75

7.5


ba (mV)

-bc (mV)

icorr (µA/cm2)

ilim. (µA/cm2)

β (icorr/ ilim.)

1000 1250 1500 1000 1250 1500 1000 1250 1500

51.5 51.3 51.5 52.4 52.4 52.3 53.0 53.2 53.2

432 427 437 618 625 608 704 744 724

63.59 69.79 82.21 90.35 102.89 115.35 142.08 144.22 148.49

173.54 191.17 159.08 208.64 219.79 209.31 262.74 236.92 278.16

0.37 0.37 0.52 0.43 0.47 0.55 0.54 0.61 0.53

Table 4.20 Corrosion Process Kinetic Parameters as a Function of Temperatures and Speeds of Rotation obtained from β-Model in Absence of Acetic Acid at pH = 8.5. Run No.

1 2 3 4 5 6 7 8 9

Temp. (°C)

pH

65

8.5

70

8.5

75

8.5


ba (mV)

-bc (mV)

icorr (µA/cm2)

ilim. (µA/cm2)

β (icorr/ ilim.)

1000 1250 1500 1000 1250 1500 1000 1250 1500

51.3 50.9 51.4 52.1 52.3 52.5 53.3 53.0 53.2

421 424 426 620 618 628 729 764 768

49.80 57.63 65.46 88.59 93.45 98.30 127.38 129.26 131.15

150.16 158.59 157.64 211.16 164.93 268.02 277.62 225.90 300.17

0.33 0.36 0.42 0.42 0.57 0.37 0.46 0.57 0.44

Tables 4.17 through 4.20, generally, show that: i.

The values of ilim. > icorr , and β ranged from (0.1- 0.9) in presence of acetic acid and (0.33-0.61) in absence of acetic acid. This indicates the mixed control corrosion mechanism.

168

Chapter Four


ii. The effect of temperature, at different pH’s, HAc concentrations and speeds of rotation, on the values of β is considerable. iii. Corrosion parameters obtained from β-Model listed in Tables 4.17 through 4.20, (i.e., limiting diffusion current density (ilim.) increased with both temperature and speed of rotation increasing. The values of icorr , bc and ba are approximately the same as the values obtained by Tafel Extrapolation and McLaughlin Methods with some deviations. This deviation may be due to the effect of mass-transport on activation process). iv. In absence of acetic acid, the values of β are lower, with average values of (0.47) and (0.44) at different pH’s, and generally, the values of β decrease as the temperature and speed of rotation increase. This reduction indicates, the formation of the protective film on the metal surface which slow down both charge and mass transfer processes, i.e., the charge transport and the mass transfer steps will be the rate determining step. Sridharan, [2009] conclude that in deaerated system, the protective film formation is accelerated by measures that restrict the transport of reaction products from the surface and inhibits the anodic reaction of AISI carbon steel 1018 in 3% (w/w) NaCl solution saturated with carbon dioxide at pH 6.5 with no pronounce effect on cathodic reaction.

4.2.3.4 API X65 Mild Steel Mass Transfer Correction Factors: Equation (2.91) can be used to determine the mass transfer correction factor (λ) at different values of β and bc. These values are listed in Appendix H at different temperatures and speeds of rotation in presence and absence of acetic acid (i.e., absence and presence the protective film formation), and shown in Figures 4.57 through 4.80.

169

Chapter Four


It is known that both activation and concentration polarizations usually occur on metal surface during corrosion process. At low reaction rates, activation polarization usually controls, while at higher reaction rates

concentration

polarization

becomes

controlling.

The

total

polarization of an electrode is the sum of the contributions of activation polarization and concentration polarization ( ηT= ηA + ηC ) Roberge, [2012].Therefore, as shown in Figures 5.57 through 5.80, that the values of λ will approach unity at low over potential and it decreases as over potential increases.

(i) API X65 Mass Transfer Correction Factors (No Protective Film Formation): Figures 4.57 through 4.68 showed the mass transfer correction factor diagrams (λ) obtained for API X65 in CO2 saturated, NaCl solution as a function of over potential, at different temperatures, solution pH values, acetic acid concentrations and speeds of rotation. In presence of acetic acid at (pH 3 and 1000 ppm HAc) and (pH 5 and 3000 ppm), generally, the values of λ increase with temperatures and speeds of rotation, and these values approach unity at low over potential. This behavior, as mentioned previously, attributed to the deposition and formation of the non-protective film (Fe3C & FeCO3) on the metal surface. Gulbrandsen et al., [1998] conclude that these films can significantly alter the corrosion process by either decreasing the corrosion rate by acting as a diffusion barrier, or increasing the corrosion by increasing the active specimen surface area by forming a conductive bridge between the counter and working electrodes.

170

Chapter Four

Results & Discussion o T = 40λ11 C; λ= =0.9986-0.0046*x+7.3259E-6*x^2 0.9654 - 0.0034 x η ; R = 0.99 λ12 = 0.999-0.0035*x+4.528E-6*x^2 T = 50 oC; λ = 0.9785 - 0.0027 x η ; R = 0.99 λ13 = 0.9997-0.0028*x+2.827E-6*x^2 T = 60 oC; λ = 0.9869 - 0.0023 x η ; R = 0.99

Mass Transfer Correction Factor λ

1.0

0.9

T = 40 o C T = 50 o C T = 60 o C

0.8

0.7

0.6

0.5

0.4 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)

Fig. 4.57 Mass Transfer Correction Factor as a Function of Over Potential in Presence of Acetic Acid at pH 3, 1000 ppm HAc & 1000 rpm.

λ21 = 1.0016-0.0024*x+3.4365E-7*x^2

o T = 40 C; λ = 1.0000 - 0.0024 x η ; R = 1.00 λ22 = 1.0002-0.001*x-8.2855E-7*x^2 o T = 50 λ = 1.0004 - 0.0012 x η ; R = 0.99 λ23 C; = 1.0001-0.0008*x-6.2296E-7*x^2 T = 60 oC; λ = 1.0029 - 0.0009 x η ; R = 0.99


1.05 1.00 0.95 0.90 0.85 0.80

T = 40 oC

0.75

o

T = 50 C T = 60 oC

0.70 0.65 0.60 0

20

40

60

80

O ve r Pote ntial

100

120

140

160

η (mV)


171

Chapter Four


o = 1.0013-0.0021*x-3.608E-7*x^2 T = 40λ31 C; λ = 1.0029 - 0.0021 x η ; R = 1.00 λ32 o = 1.0002-0.001*x-7.7733E-7*x^2 T = 50 C; λ = 1.0037 - 0.0037 x η ; R = 0.99 λ33o = 1.0001-0.0008*x-6.0614E-7*x^2 T = 60 C; λ = 1.0028 - 0.0009 x η ; R = 0.99


1.05 1.00 0.95 0.90 0.85 0.80

T = 40 o C T = 50 o C T = 60 o C

0.75 0.70 0.65 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)


λ41o=C; 1.0012-0.0021*x-3.7684E-7*x^2 T = 40 λ = 1.0030 - 0.0021 x η ; R = 1.00 λ42o= 1.0004-0.0014*x-4.9168E-7*x^2 T = 50 C; λ = 1.0026 - 0.0015 x η ; R = 0.99 λ43 = 1.0002-0.001*x-5.243E-7*x^2 T = 60 o C; λ = 1.0025 - 0.0011 x η ; R = 0.99


1.05 1.00 0.95 0.90 0.85 0.80

T = 40 o C T = 50 o C T = 60 o C

0.75 0.70 0.65 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)


172

Chapter Four


o = 1.0016-0.0024*x+2.009E-7*x^2 T = 40λ51 C; λ = 1.0007 - 0.0024 x η ; R = 1.00 λ52 = 1.0004-0.0014*x-4.4934E-7*x^2 T = 50 oC; λ = 1.0024 - 0.0014 x η ; R = 0.99 λ53 = 1-0.0004*x-5.8498E-7*x^2 T = 60 oC; λ = 1.0026 - 0.0005 x η ; R = 0.99


1.05 1.00 0.95 0.90 0.85 0.80

T = 40 oC T = 50 oC T = 60 oC

0.75 0.70 0.65 0.60 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)


T T T


1.05

=λ61 40=o1.0013-0.002*x-5.6981E-7*x^2 C; λ = 1.0039 - 0.0021 x η ; R = 1.00 =λ62 50=o1.0002-0.001*x-7.9386E-7*x^2 C; λ = 1.0038 - 0.0011 x η ; R = 0.99 λ63 = 0.9999-0.0004*x-5.2476E-7*x^2 = 60 o C; λ = 1.0023 - 0.0004 x η ; R = 0.99

1.00 0.95 0.90 0.85 0.80

T = 40 o C T = 50 o C T = 60 o C

0.75 0.70 0.65 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)

Fig. 4.62 Mass Transfer Correction Factor as a Function of Over Potential in Presence of Acetic Acid at pH 5, 3000 ppm HAc and 1500 rpm. 173

Chapter Four


λ11 =rpm 0.9986-0.0046*x+7.3259E-6*x^2 ω = 1000 ; λ = 0.9654 - 0.0034 x η ; R = 0.99 λ21 =rpm 1.0016-0.0024*x+3.4365E-7*x^2 ω = 1250 ; λ = 1.0000 - 0.0024 x η ; R = 1.00 λ31 = 1.0013-0.0021*x-3.608E-7*x^2 ω = 1500 rpm ; λ = 1.0029 - 0.0021 x η ; R = 1.00


1.1

1.0

0.9

0.8

0.7

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.6

0.5

0.4 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)

Fig. 4.63 Mass Transfer Correction Factor as a Function of Over Potential in Presence of Acetic Acid at 40 °C ,pH 3 & 1000 ppm HAc.

λ12rpm = 0.999-0.0035*x+4.528E-6*x^2 ω = 1000 ; λ = 0.9785 - 0.0027 x η ; R = 0.99 λ22rpm = 1.0002-0.001*x-8.2855E-7*x^2 ω = 1250 ; λ = 1.0040 - 0.0012 x η ; R = 1.00 λ32rpm = 1.0002-0.001*x-7.7733E-7*x^2 ω = 1500 ; λ = 1.0037 - 0.0011 x η ; R = 1.00


1.05 1.00 0.95 0.90 0.85 0.80 0.75

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.70 0.65 0.60 0.55 0

20

40

60

80

Over Potential

100

120

140

160

η (mV)

Fig. 4.64 Mass Transfer Correction Factor as a Function of Over Potential in Presence of Acetic Acid at 50 °C ,pH 3 & 1000 ppm HAc.

174

Chapter Four


λ13rpm = 0.9997-0.0028*x+2.827E-6*x^2 ω = 1000 ; λ = 0.9869 - 0.0023 x η ; R = 0.99 λ23 = 1.0001-0.0008*x-6.2296E-7*x^2 ω = 1250 rpm ; λ = 1.0029 - 0.0009 x η ; R = 0.99 λ33 = 1.0001-0.0008*x-6.0614E-7*x^2 ω = 1500 rpm ; λ = 1.0028 - 0.0009 x η ; R = 0.99


1.05 1.00 0.95 0.90 0.85 0.80 0.75

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.70 0.65 0.60 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)

Fig. 4.65 Mass Transfer Correction Factor as a Function of Over Potential in Presence of Acetic Acid and at 60 °C, pH 3 & 1000 ppm HAc.

λ41 = 1.0012-0.0021*x-3.7684E-7*x^2

ω = λ51 1000 rpm ; λ = 1.0030 - 0.0021 x η ; R = 1.00 = 1.0016-0.0024*x+2.009E-7*x^2 ω = λ61 1250= 1.0013-0.002*x-5.6981E-7*x^2 rpm ; λ = 1.0007 - 0.0024 x η ; R = 1.00 ω = 1500 rpm ; λ = 1.0039 - 0.0021 x η ; R = 1.00


1.00 0.95 0.90 0.85 0.80 0.75

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.70 0.65 0.60 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)

Fig. 4.66 Mass Transfer Correction Factor as a Function of Over Potential in Presence of Acetic Acid at 40 °C, pH 5 & 3000 ppm HAc.

175

Chapter Four


ω = 1000 ; λ = 1.0026 - 0.0015 x η ; R = 1.00 λ42 rpm = 1.0004-0.0014*x-4.9168E-7*x^2 λ52 rpm = 1.0004-0.0014*x-4.4934E-7*x^2 ω = 1250 ; λ = 1.0024 - 0.0014 x η ; R = 1.00 λ62rpm = 1.0002-0.001*x-7.9386E-7*x^2 ω = 1500 ; λ = 1.0038 - 0.0011 x η ; R = 0.99


1.00 0.98 0.96 0.94 0.92 0.90 0.88 0.86

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.84 0.82 0.80 0.78 0.76 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)


λ43rpm = 1.0002-0.001*x-5.243E-7*x^2 ω = 1000 ; λ = 1.0025 - 0.0011 x η = 1-0.0004*x-5.8498E-7*x^2 ω = 1250λ53 rpm ; λ = 1.0026 - 0.0005 x η λ63 rpm = 0.9999-0.0004*x-5.2476E-7*x^2 ω = 1500 ; λ = 1.0023 - 0.0004 x η


1.02

; R = 1.00 ; R = 1.00 ; R = 0.99

1.00 0.98 0.96 0.94 0.92 0.90

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.88 0.86 0.84 0.82 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)


176

Chapter Four


(ii) API X65 Mass Transfer Correction Factors (Under Protective Film Formation): Figures 4.69 through 4.80 show the mass transfer correction factor (λ) diagrams in presence of protective film formation, at different temperatures, solution pH values and speeds of rotation. In absence of acetic acid at pH 8.5, generally, the values of λ increase with temperatures and speeds of rotation, and these values approach unity at low over potential. The same behavior observed with absence of acetic acid at pH 7.5 to less extent. This behavior, as mentioned previously, attributed to the deposition and formation of the protective film on the metal surface. This deposition of the protective film formation is accelerated by measures that restrict the transport of reaction products from the surface, and inhibits both the anodic and cathodic reactions. Anodic reaction was assumed to be always under chargetransfer control, while the cathodic reaction is a mixed control, so that, the deposition of the protective film will slow down the charge-transfer step. This step will control the corrosion process more than the masstransport step and the value of λ approach unity (i.e., the correction factor becomes less important at low over potential and as the temperature and speed of rotation increased).

177

Chapter Four

Results & Discussion o = 1.0012-0.002*x-3.3038E-7*x^2 T = 65λ71 C; λ = 1.0026 - 0.0021 x η ; R = 1.00 λ72o C; = 1.0005-0.0016*x+4.3269E-8*x^2 T = 70 λ = 1.0003 - 0.0016 x η ; R = 1.00 λ73o = 1.0003-0.0018*x+5.7863E-7*x^2 T = 75 C; λ = 0.9977 - 0.0017 x η ; R = 1.00


1.00

0.95

0.90

0.85

0.80

T = 65 o C T = 70 o C T = 75 o C

0.75

0.70

0.65 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)

Fig. 4.69 Mass Transfer Correction Factor as a Function of Over Potential in Absence of Acetic Acid at pH 7.5 & 1000 rpm.

o T = 65 λ81 C;=λ1.0012-0.002*x-3.394E-7*x^2 = 1.0027 - 0.0021 x η ; R = 1.00 λ82 o = 1.0005-0.0018*x+2.6145E-7*x^2 T = 70λ83 C; λ = 0.9993 - 0.0017 x η ; R = 1.00 = 1.0002-0.0019*x+8.9802E-7*x^2 T = 75 o C; λ = 0.9962 - 0.0018 x η ; R = 0.99


1.00

0.95

T = 65 o C T = 70 o C T = 75 o C

0.90

0.85

0.80

0.75

0.70

0.65 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)


178

Chapter Four


λ91 = 1.0013-0.0028*x+1.591E-6*x^2

T = λ92 65 o=C;1.0005-0.0021*x+9.2286E-7*x^2 λ = 0.9941 - 0.0026 x η ; R = 1.00 T = λ93 70 o=C;1.0003-0.0017*x+4.9945E-7*x^2 λ = 0.9963 - 0.0020 x η ; R = 0.99 T = 75 o C; λ = 0.9980 - 0.0016 x η ; R = 1.00


1.00 0.95 0.90 0.85 0.80 0.75

T = 65 o C T = 70 o C T = 75 o C

0.70 0.65 0.60 0.55 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)


T = λ101 65 o C; λ = 1.0043 - 0.0020 x η ; R = 1.00 = 1.0011-0.0019*x-7.0299E-7*x^2 = 1.0004-0.0016*x-4.0251E-8*x^2 T = λ102 70 o C; λ = 1.0006 - 0.0016 x η ; R = 1.00 T =λ103 75 o=C;1.0003-0.0015*x+1.0008E-7*x^2 λ = 0.9998 - 0.0015 x η ; R = 1.00


1.00

0.95

0.90

0.85

0.80

T = 65 o C T = 70 o C T = 75 o C

0.75

0.70

0.65 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)

Fig. 4.72 Mass Transfer Correction Factor as a Function of Over Potential in Absence of Acetic Acid at pH 8.5 & 1000 rpm. 179

Chapter Four

Results & Discussion o = 1.0012-0.0021*x-3.5542E-7*x^2 T = 65λ111 C; λ = 1.0028 - 0.0021 x η ; R = 1.00 λ112 o = 1.0005-0.0021*x+1.0105E-6*x^2 T = 70 C; λ = 0.9959 - 0.0020 x η ; R = 0.99 λ113 = 1.0003-0.0017*x+6.3937E-7*x^2 T = 75 o C; λ = 0.9974 - 0.0016 x η ; R = 1.00


1.00

0.95

0.90

0.85

0.80

T = 65 o C T = 70 o C T = 75 o C

0.75

0.70

0.65 0

20

40

60

80

Over Potential

100

120

140

160

η (mV)


T = 65 oλ121 C; λ == 1.0013-0.0023*x+2.7448E-7*x^2 1.0001 - 0.0023 x η ; R = 1.00 λ122 = 1.0004-0.0014*x-3.2508E-7*x^2 T = 70 oλ123 C; λ = 1.0018 - 0.0014 x η ; R = 1.00 = 1.0002-0.0013*x-1.3712E-8*x^2 T = 75 o C; λ = 1.0003 - 0.0013 x η ; R = 1.00


1.05 1.00 0.95 0.90 0.85 0.80

T = 65 o C T = 70 o C T = 75 o C

0.75 0.70 0.65 0.60 0

20

40

60

80

O ve r Pote ntial

100

120

140

160

η (mV)


180

Chapter Four

Results & Discussion ω ω ω

= 1.0012-0.002*x-3.3038E-7*x^2 =λ71 1000 rpm ; λ = 1.0026 - 0.0021 x η ; R = 1.00 = 1.0012-0.002*x-3.394E-7*x^2 = λ81 1250 rpm ; λ = 1.0027 - 0.0021 x η ; R = 1.00 = 1.0013-0.0028*x+1.591E-6*x^2 =λ91 1500 rpm ; λ = 0.9941 - 0.0026 x η ; R = 0.99


1.00 0.95 0.90 0.85 0.80 0.75

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.70 0.65 0.60 0.55 0

20

40

60

80

Over Potential

100

120

140

160

η (mV)

Fig. 4.75 Mass Transfer Correction Factor as a Function of Over Potential in Absence of Acetic Acid at 65°C & pH 7.5

= 1.0005-0.0016*x+4.3269E-8*x^2 ω = 1000λ72 rpm ; λ = 1.0003 - 0.0016 x η ; R = 1.00 = 1.0005-0.0018*x+2.6145E-7*x^2 ω = 1250λ82 rpm ; λ = 1.9993 - 0.0017 x η ; R = 1.00 λ92 = 1.0005-0.0021*x+9.2286E-7*x^2 ω = 1500 rpm ; λ = 0.9963 - 0.0020 x η ; R = 0.99


1.00

0.95

0.90

0.85

0.80

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.75

0.70

0.65 0

20

40

60

80

O ve r Pote ntial

100

120

140

160

η (mV)

Fig. 4.76 Mass Transfer Correction Factor as a Function of Over Potential in Absence of Acetic Acid at 70°C & pH 7.5.

181

Chapter Four


ω = 1000λ73 rpm= ;1.0003-0.0018*x+5.7863E-7*x^2 λ = 0.9977 - 0.0017 x η ; R = 1.00 ω = 1250λ83 rpm= ;1.0002-0.0019*x+8.9802E-7*x^2 λ = 0.9962 - 0.0018 x η ; R = 1.00 ω = 1500λ93 rpm= ;1.0003-0.0017*x+4.9945E-7*x^2 λ = 0.9980 - 0.0016 x η ; R = 1.00


1.00 0.98 0.96 0.94 0.92 0.90 0.88 0.86 0.84

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.82 0.80 0.78 0.76 0.74 0.72 0

20

40

60

80

Over Potential

100

120

140

160

η (mV)


ω = 1000λ101 rpm =; 1.0011-0.0019*x-7.0299E-7*x^2 λ = 1.0043 - 0.0020 x η ; R = 1.00 ω = 1250λ111 rpm =; 1.0012-0.0021*x-3.5542E-7*x^2 λ = 1.0028 - 0.0021 x η ; R = 1.00 ω = 1500λ121 rpm=; 1.0013-0.0023*x+2.7448E-7*x^2 λ = 1.0001 - 0.0023 x η ; R = 1.00


1.00 0.95 0.90 0.85 0.80

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.75 0.70 0.65 0.60 0

20

40

60

80

Over Potential

100

120

140

160

η (mV)

Fig. 4.78 Mass Transfer Correction Factor as a Function of Over Potential in Absence of Acetic Acid at 65°C & pH 8.5 182

Chapter Four


ω = 1000λ102 rpm=;1.0004-0.0016*x-4.0251E-8*x^2 λ = 1.0006 - 0.0016 x η ; R = 1.00 ω = 1250λ112 rpm= ;1.0005-0.0021*x+1.0105E-6*x^2 λ = 0.9959 - 0.0020 x η ; R = 0.99 ω = 1500λ122 rpm=;1.0004-0.0014*x-3.2508E-7*x^2 λ = 1.0018 - 0.0014 x η ; R = 1.00


1.00

0.95

0.90

0.85

0.80

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.75

0.70

0.65 0

20

40

60

80

O ver Potential

100

120

140

160

η (mV)

Fig. 4.79 Mass Transfer Correction Factor as a Function of Over Potential in Absence of Acetic Acid at 70°C and pH 8.5

λ103 ω = 1000 rpm= ;1.0003-0.0015*x+1.0008E-7*x^2 λ = 0.9998 - 0.0015 x η ; R = 1.00 λ113 ω = 1250 rpm= ;1.0003-0.0017*x+6.3937E-7*x^2 λ = 0.9974 - 0.0016 x η ; R = 1.00 λ123 ω = 1500 rpm= ;1.0002-0.0013*x-1.3712E-8*x^2 λ = 1.0003 - 0.0013 x η ; R = 1.00


1.00 0.98 0.96 0.94 0.92 0.90 0.88 0.86 0.84

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.82 0.80 0.78 0.76 0.74 0

20

40

60

80

Over Potential

100

120

140

160

η (mV)


183

Chapter Four


4.2.3.5 API X65 Mild Steel Polarization Resistances: The approach in this investigation was suggested here to analyze the values of polarization resistance by using the coefficients of β –Model Korobove and Medvedeva, [2000]. Let us first define the concepts of polarization resistance. The polarization resistance is a measure of a metal to undergo corrosion in a certain environment, it is a slope of the potential-current (E-i) curve at the corrosion potential, i.e.: -

 E  Rp     i  E Ecorr

...(4.26)

The shape of this curve around the corrosion potential, and hence R p, is determined by the sum of the partial E-i curves of both anodic and cathodic half reactions Wagner and Traud, [1938] and Stern and Geary, [1957]. The shape of both curves, in turn, depends on the respective kinetic parameters of their reaction. The current-potential curve of corroding metal is rather complex non-linear Equation; hence a general analytical solution for Rp is equally complex. Under some conditions, the E-i relation reduces to a simple form, and hence a simple analytical solution for Rp can be obtained. However, the (E-i) relation can be represented by Wagner and Traud Equation as,   i  iCorr. exp   ba 

     exp     bc

  

(2.68)

Where η is over potential (i.e., E-Ecorr). A derivative form of Equation (2.68) is the Stern and Geary Equation, which can be written as; ba bc    Rp      i  0 2.303icorr b a  b c 

(2.69)

184

Chapter Four


This Equation assumed that the applied current is a linear function of electrode potential (about 10 mV around corrosion potential), and slope of this linear relationship is Rp. Equation (2.69) will have high percent of error because of the non-linearity of (η-i) relationship near the corrosion potential. According (

η ) or ( i

Resistance [1973] (

to

Stern,

the

quotient

) is called the polarization resistance by analogy to ( has

Ε ) from conventional electricity. Clerbois Pourbaix, Ι

suggested

that

the

inverse

of

), be called the “corrodance” by obvious analogy to the inverse of

electrical resistance, which is called the “conductance”. Recently, Mansfeld Mansfeld, [2005] stated that, as new experimental technique mature and are more commonly used, various misconception and misnomers are sometimes perpetuated, which, when used often enough, are considered to be truth. Some of these myths, misconception, and misnomers are the linear polarization technique. There exists the polarization resistance (Rp) technique and linear polarization technique, but there is no linear polarization resistance (LPR) technique. As pointed out many years ago David and James, [1998] and Robert et al., [2003], polarization curves recorded in the vicinity of the corrosion potential (Ecorr) are linear only in the exceptional case that the anodic (ba) and cathodic (bc) Tafel slopes are equal, i.e., bc = ba. In general case there can be appreciable curvature at Ecorr. It is sometimes mentioned that there is serious drawback in the Rp 

technique, because accurate values of the parameter B    

185

 ba bc , 2.303b a  b c  

Chapter Four


which are required to convert experimental values of R p in to corrosion 

 

current density  icorr   are not known for the conditions and the time Rp   of an experiment. This argument ignores the fact that, at least in corrosion monitoring, experimental values of Rp can be used to detect significant changes in corrosion rates that require, for example, the protective film formation. The term polarization resistance (Rp) must be not confused with resistance polarization (R). Rp represents the charge transfer resistance resulting from the separation of charges across the solid interface to the outside edge of the double layer. While R represents the resistance through the bulk solution (Rs) and the resistance of any films present on the working electrode surface (Rf) (i.e., R = Rs+Rf) Tretheway and Chamberlain, [1996]. The approach that suggested here was to use the values of the second coefficient of β-Model to evaluate the value of Rp, i.e., by using Equation (2.95);

 1   1   C1  2.303 icorr     1     b a   b c  

… (2.95)

 1   . This analysis may take in to  R  p

It is clear that when   0 , C1  

account the nonlinearly of polarization curve near the corrosion potential. The polarization resistance of API X65 mild steel in CO2 saturated, NaCl solution in presence and absence of acetic acid at different conditions was studied using the values of the second coefficient (C1) of β-Model Korobove and Medvedeva, [2000] results given in Appendix E.

186

Chapter Four


(i) API X65 Mild Steel Polarization Resistances (No Protective Film Formation): Table 4.21 shows the values of polarization resistance (Rp) obtained for API X65 in CO2 saturated, NaCl solution in presence of acetic acid at different temperatures and speeds of rotation. Table 4.21 The Effect of Temperature and Speed of Rotation on Polarization Resistance of API X65Mild Steel in CO2 Saturated, 3.5 wt% NaCl Solution in Presence of Acetic Acid (Absence of Protective Film Formation). pH

HAc Conc. (ppm)

3

1000

5

3000

Speed of Rotation (rpm) 1000 1250 1500 1000 1250 1500

Polarization Resistance, Rp (Ω.cm2) x 10-3 40 ºC

50 ºC

60 ºC

0.1087 0.0961 0.0887 0.1335 0.1255 0.1196

0.0866 0.0793 0.0652 0.1256 0.1176 0.1088

0.0457 0.0406 0.0399 0.1004 0.0992 0.0984

* The polarization resistance is x1000 Figures 4.81 through 4.84 show the effect of both temperature and speed of rotation on the polarization resistance in absence of the protective film formation at different conditions. It is clear that: The values of Rp in presence of acetic acid at (pH 5 & 3000 ppm HAc) are higher than the values at (pH 3 & 1000 ppm HAc) at different temperature. This indicate the high efficiency of the non protective film formation at (pH 5 & 3000 ppm HAc) to decrease the corrosion rate by acting as a diffusion barrier and to restrict the transport of reaction products of the corrosion of both charge and mass transfer processes from the surface.

187

Chapter Four

Results & Discussion 0.12 0.11

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

Rp x 10 -3 (ohm.cm 2)

0.10 0.09 0.08 0.07 0.06 0.05 0.04 0.03 38

40

42

44

46

48

50

52

54

56

58

60

62

o

Temperature ( C)

Fig. 4.81 The Effect of Temperature on Polarization Resistance of API X65 Mild Steel in Presence of Acetic Acid in CO2 Saturated,3.5 wt% NaCl Solution at pH 3 & HAc Conc. 1000 ppm.

0.12

T = 40 o C T = 50 o C T = 60 o C

0.11

Rp x 10 -3 (ohm.cm2)

0.10 0.09 0.08 0.07 0.06 0.05 0.04 0.03 900

1000

1100

1200

1300

1400

1500

1600


Fig. 4.82 The Effect of Speed of Rotation on Polarization Resistance of API X65 Mild Steel in Presence of Acetic Acid in CO2 Saturated, 3.5 wt% NaCl Solution at pH 3 & HAc Conc. 1000 ppm.

188

Chapter Four

Results & Discussion 0.135

0.130

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

Rp x 10 -3 (ohm.cm 2)

0.125

0.120

0.115

0.110

0.105

0.100

0.095 38

40

42

44

46

48

50

52

54

56

58

60

62

Temperature (oC)

Fig. 4.83 The Effect of Temperature on Polarization Resistance of API X65 Mild Steel in Presence of Acetic Acid in CO2 Saturated,3.5 wt% NaCl Solution at pH 5 & HAc Conc. 3000 ppm.

0.135

T = 40 o C T = 50 o C T = 60 o C

0.130


0.125

0.120

0.115

0.110

0.105

0.100

0.095 900

1000

1100

1200

1300

1400

1500

1600


Fig. 4.84 The Effect of Speed of Rotation on Polarization Resistance of API X65 Mild Steel in Presence of Acetic Acid in CO2 Saturated, 3.5 wt% NaCl Solution at pH 5 & HAc Conc. 3000 ppm. 189

Chapter Four


(ii) API X65 Mild Steel Polarization Resistances (Under Protective Film Formation): Table 4.22 shows the values of polarization resistance (Rp) obtained for API X65 in CO2 saturated, NaCl solution in absence of acetic acid at different temperatures and speeds of rotation. Table 4.22 The Effect of Temperature and Speed of Rotation on Polarization Resistance of API X65Mild Steel in CO2 Saturated, 3.5 wt% NaCl Solution in Absence of Acetic Acid (Presence of Protective Film Formation). Speed of Rotation (rpm) 1000 1250 1500 1000 1250 1500

pH

7.5

8.5

Polarization Resistance, Rp (Ω.cm2) x 10-3 65 ºC

70 ºC

75 ºC

0.3803 0.3454 0.2884 0.4869 0.4152 0.3668

0.2645 0.2315 0.2047 0.2684 0.2522 0.2448

0.1677 0.1647 0.1610 0.1891 0.1834 0.1832

* The polarization resistance is x1000 Figures 4.85 through 4.88 showed the effect of temperature, solution pH and speed of rotation on the polarization resistance in presence of the protective film formation at different conditions. It is clear that: i.

The polarization resistance values in absence of acetic acid are larger than the polarization resistance values in presence of acetic acid, due to the formation of the protective film.

ii.

The values of Rp in absence of acetic acid at pH 8.5 are higher than the values at pH 7.5, indicating the resistance to the corrosion in presence of protective film formation at pH 8.5 is more than the resistance in presence of protective film formation at pH 7.5. This reflects the high efficiency of the protective film formation at pH 8.5 to restrict the transport of reaction products of the corrosion of both 190

Chapter Four


charge and mass transfer processes from the surface. Hernandez et al., [2012] studied the formation effect of the protective dense FeCO3 scale layer on the corrosion of API X70 pipeline steel in 3 % NaCl solution saturated with CO2 and deaerated in absence of acetic acid at pH 6.6 under controlled turbulent flow conditions. He found that the values of Rp for the carbon steel rotating cylinder electrode (RCE) increase with decreasing the temperature and the rotation rate. The values of Rp are deduced for the working electrode from the slopes of   .  i 

the straight lines obtained from potential-current plots 

0.40 0.38 0.36

R p x 10 -3 (ohm.cm2)

0.34

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.32 0.30 0.28 0.26 0.24 0.22 0.20 0.18 0.16 0.14 64

66

68

70

72

74

76

o

Temperature ( C)

Fig. 4.85 The Effect of Temperature on Polarization Resistance of API X65 Mild Steel in Absence of Acetic Acid in CO2 Saturated, 3.5 wt% NaCl Solution at pH 7.5.

191

Chapter Four

Results & Discussion 0.40 0.38

T = 65 o C T = 70 o C T = 75 o C

0.36

Rp x 10 -3 (ohm.cm 2)

0.34 0.32 0.30 0.28 0.26 0.24 0.22 0.20 0.18 0.16 0.14 900

1000

1100

1200

1300

1400

1500

1600


Fig. 4.86 The Effect of Speed of Rotation on Polarization Resistance of API X65 Mild Steel in Absence of Acetic Acid in CO2 Saturated,3.5 wt% NaCl Solution at pH 7.5.

0.50


0.45

ω = 1000 rpm ω = 1250 rpm ω = 1500 rpm

0.40

0.35

0.30

0.25

0.20

0.15 64

66

68

70

72

74

76

Temperature (oC)

Fig. (4.87) The Effect of Temperature on Polarization Resistance of API X65 Mild Steel in Absence of Acetic Acid in CO2 Saturated, 3.5 wt% NaCl Solution at pH 8.5.

192

Chapter Four

Results & Discussion 0.50

T = 65 o C T = 70 o C T = 75 o C

0.40

0.35

Rp x 10

-3

(ohm.cm 2)

0.45

0.30

0.25

0.20

0.15 900

1000

1100

1200

1300

1400

1500

1600


Fig. (4.88) The Effect of Speed of Rotation on Polarization Resistance of API X65 Mild Steel in Absence of Acetic Acid in CO2 Saturated,3.5 wt% NaCl Solution at pH 8.5

4.2.3.6 Evaluation of Corrosion Parameters: Values of the corrosion parameters (icorr and

(

))

from β-

Model [Korobove and Medvedeva, 2000] results are given in Tables 4.17 through 4.20. Corrosion current densities vs. Re in presence and absence of acetic acid at different temperatures and pH values are shown in Figures 4.89 through 4.92. The relations can be represented by straight lines with slopes (△ icorr / △ Re)

T, pH

shown in Table 4.23. Noting that the total

cathodic current density is the summation of the hydrogen evolution reaction and the hydrogen limiting current densities.

193

Chapter Four


Table 4.23 (△ icorr / △ Re) in Presence and Absence of Acetic Acid at Different Temperatures and pH Values. Temp. (°C) 40 50 60 40 50 60 65 70 75 65 70 75

pH

HAc Conc. (ppm)

3

1000

5

3000

7.5

Blank Solution

8.5

Blank Solution

△ icorr / △ Re (μA/cm2) 3.466 5.084 4.103 0.844 1.520 0.725 0.758 0.952 0.228 0.637 0.369 0.133

At the solution pH’s (3 and 5) and HAc acid concentration (10003000 ppm) respectively, (△ icorr / △ Re) increases with increasing temperature at 50°C, then decreases at 60°C and at the solution pH = 7.5 in absence of acetic acid it increases with increasing temperature at 70°C, then decreases at 75 °C, while at the solution pH = 8.5 it decreases with increasing temperature.

194

Chapter Four

Results & Discussion = 81.8521+3.4656*x T = 40 o C icorr ; i corrat=40oC 81.8521 + 3.4656 x Re; R = 0.99 icorr at 50oC = 49.989+5.0843*x T = 50 o Cicorr ; i corrat=60oC 49.9890 + 5.0843 x Re; R = 0.99 = 309.5051+4.1033*x T = 60 o C ; i corr = 309.5051 + 4.1033 x Re;R = 0.99

650 600

icorr ( µA / cm2 )

550 500 450 400 350 300

T = 40 o C T = 50 o C T = 60 o C

250 200 150 30

35

40

45

50

55

60

65

70

Re x 1000

Fig. 4.89 Variation of Corrosion Current of API X65 Mild Steel with Re in CO2 Saturated,3.5 wt% NaCl Solutions in Presence of Acetic Acid at pH 3 and 1000 ppm HAc.

at40oC = 140.457+0.8436*x T = 40 oicorr C ; icorr = 140.4570 + 0.8436 x Re; R = 1.00 icorr at50oC = 117.966+1.5204*x

o T = 50 icorr C ; iatcorr = 117.9660 + 1.5204 x Re; R = 1.00 60oC = 195.2766+0.7245*x

T = 60 oC ; icorr = 195.2766 + 0.7245 x Re; R = 1.00

250 240

i corr ( µA / cm 2 )

230 220 210 200 190 180

o

T = 40 C o

T = 50 C

170

o

T = 60 C 160 30

35

40

45

50

55

60

65

70

Re x 1000

Fig. 4.90 Variation of Corrosion Current of API X65 Mild Steel with Re in CO2 Saturated,3.5 wt% NaCl Solutions in Presence of Acetic Acid at pH 5 and 3000 ppm HAc.

195

Chapter Four

Results & Discussion o 65 = oC26.1743 = 26.1743+0.7575*x T = 65 icorrat C ; icorr + 0.7575 x Re; R = 0.99

icorr at 70oC = 40.3408+0.9516*x

o T = 70 icorr C ; at icorr = 40.3408 + 0.9516 x Re; R = 1.00 75oC = 129.285+0.2275*x

T = 75 oC ; icorr = 129.2850 + 0.2275 x Re;R = 0.99

160 150 140

icorr ( µA / cm2 )

130 120 110 100 90 80

o

T = 65 C o

70

T = 70 C

60

T = 75 C

o

45

50

55

60

65

70

75

80

85

90

Re x 1000

Fig. 4.91 Variation of Corrosion Current of API X65 Mild Steel with Re in CO2 Saturated, 3.5 wt% NaCl Solutions in Absence of Acetic Acid at pH 7.5.

65oC = 18.4994+0.6367*x T = 65 oicorr C ; iat corr = 18.4994 + 0.6367 x Re; R = 1.00 icorr at 70oC = 69.1787+0.3694*x icorr at 75oC = 119.858+0.1333*x

T = 70 o C ; icorr = 69.1787 + 0.3694 x Re; R = 1.00 T = 75 o C ; icorr = 119.8580 + 0.1333 x Re;R = 1.00

140 130

icorr ( µA / cm2 )

120 110 100 90 80 70 60

T = 65 o C T = 70 o C T = 75 o C

50 40 45

50

55

60

65

70

75

80

85

90

Re x 1000

Fig. 4.92 Variation of Corrosion Current of API X65 Mild Steel with Re in CO2 Saturated, 3.5 wt% NaCl Solutions in Absence of Acetic Acid at pH 8.5.

196

Chapter Four


The speed of rotation of the corrosive medium increases the corrosion rate only if the process is controlled by concentration polarization. The effect of speed of rotation is to increase the rate of mass transport. Therefore in deaerated acids, carbon steel corrodes at the same rate under either low or high-speed of rotation conditions only if the process is controlled by activation polarization Nervana, [2010]. Therefore the corrosion current and the limiting diffusion current of hydrogen are function of speed of rotation in deareated solutions. The corrosion rate increased linearly with Re in the range of speeds of rotation studied for all conditions in absence and presence of the protective film, as shown in Figures 4.89 through 4.92 and Table 4.23.

4.2.3.7 Effect of Environmental Conditions on Hydrogen Limiting Current: Since hydrogen reduction reaction is diffusion controlled, the rate can be calculated using mass transfer correlations that relates the ⁄ ), the Reynolds number (

Sherwood number (

) and the Schmidt number (

⁄

). The relationship takes

the form: (

)

Where c, x and y are constants depending on system hydrodynamic characteristics. The mass transfer coefficient can be calculated from the following Equation: (

)

At the limiting current density, the surface concentration is normally equal to zero. Therefore the mass transfer coefficient for hydrogen reduction is given by: (

)

(

)

( 197

)

Chapter Four

where


is the dissolved hydrogen ion bulk concentration in

.

(i) Effect of Re: Log hydrogen limiting current density vs. log Re in presence and absence of acetic acid (absence and presence of the protective film formation) at a given temperature and pH value are shown in Figures 4.93 through 4.96. Since for a given temperature ρ, µ, D are constant at different pH values and Cb depend on pH, the slope (△log

(

)

/ △log

Re) is equal to the power x in Equation (4.27). The results are shown in Table 4.24. Also the data in these Figures increases with increasing temperatures in presence and absence of acetic acid. The dependence of

(

)

on Re in presence and absence of

acetic acid, generally decreases with increasing solution pH in the range 3 to 8.5 and increased with increasing of temperature. Table 4.24 (△log ( ) / △log Re) Sc in Presence and Absence of Acetic Acid at Different Temperatures and pH Values. Temp. (°C) 40 50 60 40 50 60 65 70 75 65 70 75

pH

HAc Conc. (ppm)

3

1000

5

3000

7.5

Blank Solution

8.5

Blank Solution

198

△log ( △log Re 0.655 1.402 1.563 0.279 1.028 1.712 0.185 0.219 1.121 0.121 0.529 0.681

)

/

Chapter Four


3.6

logilimH at +40C = -9.5825+ 2.6552*x T = 40 o C ; log id lim.(H ) = -9.5825 + 0.6552 x log Re; R = 0.99 logilimH at +50C = -15.9535+ 4.0201*x T = 50 o C ; log id lim.(H 1.4021 x log Re; R = 0.99 ) = -15.9535 logilimH at 60C = -13.7559++3.5634*x T = 60 o C ; log id lim.(H+) = -13.7559 + 1.5634 x log Re; R = 1.00

Log i d lim.(H+ ) ( µA / cm2 )

3.4

3.2

3.0

2.8

T = 40 o C T = 50 o C T = 60 o C

2.6

2.4

2.2 4.50

4.55

4.60

4.65

4.70

4.75

4.80

4.85

Log ( Re )

Fig. 4.93 Variation of Limiting Diffusion Current of Hydrogen on API X65Mild Steel in Presence of Acetic Acid in CO2 Saturated, 3.5 wt% NaCl Solutions at pH 3 & HAc Conc. 1000 ppm.

d + log iilimH = 1.4+0.2787*x T = 40 o C ; log + 0.2787 x log Re; R = 0.99 lim.(H at40C ) = 1.4146 + C = -2.0125+1.028*x at50 T = 50 o C ;logIlimH log id lim.(H ) = -2.0125 + 1.0280 x log Re; R = 1.00

logilimH at 60C = -9.7153+2.712*x

3.5

T = 60 o C ; log id lim.(H+) = -9.7153 + 1.7120 x log Re; R = 1.00

Log idlim.(H +) ( µA / cm2 )

3.4 3.3 3.2 3.1

T = 40 o C T = 50 o C T = 60 o C

3.0 2.9 2.8 2.7 2.6 4.50

4.55

4.60

4.65

4.70

4.75

4.80

4.85

Log ( Re )

Fig. 4.94 Variation of Limiting Diffusion Current of Hydrogen on API X65 Mild Steel in Presence of Acetic Acid in CO2 Saturated, 3.5 wt% NaCl Solutions at pH 5 & HAc Conc. 3000 ppm.

199

Chapter Four


2.44

logilimH at+65C = 1.3571+0.1848*x T = 65 o C ; log id lim.(H ) = 1.3571 + 0.1848 x log Re ; R = 0.99 log ilimH2 at70C = 2.2316+0.0199*x d + T = 70 o C ; log i = 2.2316 + 0.2199 x log Re ; R = 0.99 lim.(Hat75C ) log ilimH2 = 1.8274+0.1208*x o d + T = 75 C ; log i lim.(H ) = 1.8274 + 1.1208 x log Re ; R = 0.99

2.42

Log id lim. (H+ ) ( µA /cm 2 )

2.40 2.38 2.36 2.34 2.32 2.30 2.28 2.26

T = 65 o C T = 70 o C T = 75 o C

2.24 2.22 2.20 4.66 4.68 4.70 4.72 4.74 4.76 4.78 4.80 4.82 4.84 4.86 4.88 4.90 4.92 4.94 4.96

Log (Re)

Fig. 4.95 Variation of Limiting Diffusion Current of Hydrogen on API X65 Mild Steel in Absence of Acetic Acid in CO2 Saturated, 3.5 wt% NaCl Solutions at pH 7.5.

2.46

d + log iilimh2 at65C = 1.6115+0.1206*x T = 65 o C ; log lim.(H ) = 1.6115 + 0.1206 x log Re ; R = 0.99 log IlimH2 at o d + 70C = -0.2222+0.529*x T = 70 C ; log -0.2222 + 0.5291 x log Re ; R = 1.00 lim.(H ) = log iIlim H2 at 75C = 1.7176+0.146*x T = 75 o C ; log id lim.(H+) = 1.7176 + 0.6810 x log Re ; R = 1.00

2.44 2.42

Log idlim. (H +) ( µA /cm2 )

2.40 2.38 2.36 2.34 2.32 2.30

T = 65 o C T = 70 o C T = 75 o C

2.28 2.26 2.24 2.22 2.20 2.18

2.16 4.66 4.68 4.70 4.72 4.74 4.76 4.78 4.80 4.82 4.84 4.86 4.88 4.90 4.92 4.94 4.96

Log (Re)

Fig. 4.96 Variation of Limiting Diffusion Current of Hydrogen on API X65 Mild Steel in Absence of Acetic Acid in CO2 Saturated, 3.5 wt% NaCl Solutions at pH 8.5. The corrosive medium speed of rotation increases also the limiting diffusion reduction current of hydrogen which would shift to more 200

Chapter Four


negative values of the corrosion potential and polarization curves by contributing to the cathodic reaction and thus increasing the dissolution rate of the metal. Table 4.15 shows the shifting to more negative (to more active) of the corrosion potential with speed of rotation. The limiting diffusion reduction current of hydrogen increased linearly with Re in the range of speeds of rotation studied for all conditions, as shown in Figures 4.93 through 4.96. The effect of Re on the limiting diffusion currents of hydrogen, plotted on log-log scales, for constant temperature and a given pH value. Since the physical properties are constants at a given temperature, then [

△

(

△

)

]

[

The slopes (△log

△ △

(

]

(

)

)

/ △log Re) in absence of the protective

film formation, ranged from 0.65 to 1.56 in the pH 3 and ranged from 0.27 to 1.71 in the pH 5, but in presence of the protective film formation, ranged from 0.18 to 1.12 in the pH 7.5 and ranged from 0.12 to 0.68 in the pH 8.5, as shown in Table 4.24. For fully developed mass-transfer conditions, the Chilton and Colburn [analogy shows that Sh changes with Re0.8, which is proved by several investigators in deaerated mass transfer and turbulent flow conditions Mahinpey, [2001]. Bahar et al., [2013] derived a correlation empirical Equation for mass-transfer of corrosion behavior of carbon steel under flowing condition using rotating cylinder electrode as a working electrode in which ilim. changes with Re0.011. For mass-transfer in the entrance section (developing concentration boundary layer) less values for the power of Re were obtained with the most appropriate values at about 0.86 Anderko et al., [2001] and about 0.59 Wang and Nesic, [2003].

201

Chapter Four


(ii) Effect of Sc: Log

vs. log

at a given Re in presence and absence of acetic

acid and pH values are shown in Figures 4.97 through 4.100. The mass transfer coefficient (

) is calculated according to

Equation (4.29) for the determination of Sherwood number ( ⁄ ). Since the system was open and constant pH of solution, the solubility of hydrogen in water at different temperatures (total pressure of CO2 and water vapor 1 atm.) Seo et al., [2012] can be used as the bulk concentration. The diffusion coefficients ( ) of hydrogen ion in water at different temperatures can be calculated from the Stokes-Einstein following Equation Uhlig, [2011]: ( Where

) (

(

)

is the diffusion coefficient at a reference temperature

is the dynamic viscosity in

(

) and

(

)

,

is the dynamic viscosity at

a reference temperature. At 20°C, the dynamic viscosity (

)

of water is

) David, [2007] and the diffusion coefficient for H+ ions is

equal 9.31 x 10-9 m2/s Li, [2011].

For electrolytes the diffusivity can be calculated from the following Equation John and Karen, [2004]: ( Where

and ) and

)

are the diffusivities in electrolyte solution and in pure

water, respectively, (

(

√ )

is the concentration of the electrolyte solution in

is an empirical constant.

The water density in kg/m3 is found from the following Equation Naftz et al., [2011]: (

202

)

Chapter Four


And water viscosity in kg / (m s) is found from the following Equation David, [2007]: (

)

(

)

(

)

The density and viscosity of the electrolyte seawater solution (saltwater) were taken to be the same as those for water Sharqawy et al., [2010] at the respective temperature. Hameed et al., [2009] used the density and viscosity of water for seawater (3.5 % W/V NaCl solution). The numerical values of the physical properties used for the present calculations of

and

are shown in Tables 4.25 and 4.26.

log ShatRe3000 = 0.2784+0.6835*x

2.1

Re = 30000; log Sh = 0.2784 + 0.6835 x logSc; R = 0.99 logShatRe50000 = -0.2962+1.4188*x Re = 50000;logShatRe70000 log Sh = 0.2962 + 1.4188 x logSc; R = 1.00 = 0.1698+1.1422*x Re = 70000; log Sh = 0.1698 + 1.1422 x logSc; R = 1.00

2.0 1.9

Log Sh

1.8

Re = 30000 Re = 60000 Re = 90000

1.7 1.6 1.5 1.4 1.3 1.2 1.1 1.30

1.35

1.40

1.45

1.50

1.55

1.60

1.65

1.70

Log S c

Fig. 4.97 Sherwood Number vs. Schmidt Number for a Given Re in Presence of Acetic Acid in CO2 Saturated, 3.5 wt% NaCl Solutions at pH 3 & HAc Acetic Conc. 1000 ppm.

203

Chapter Four


log ShatRe=30000 = 3.344+0.1194*x

Re = 30000; log Sh = 3.3440 + 0.1194 * logSc; R = 0.99 logShatRe=50000 = 1.5292+1.4155*x Re = 50000; log Sh = 1.5292 + 1.4155 * logSc; R = 1.00 logShatRe=70000 = 1.6557+1.4154*x Re = 70000; log Sh = 1.6557 + 1.4154 * logSc; R = 0.99

4.1 4.0

Re = 30000 Re = 60000 Re = 90000

3.9

Log Sh

3.8 3.7 3.6 3.5 3.4 3.3 1.30

1.35

1.40

1.45

1.50

1.55

1.60

1.65

1.70

Log Sc

Fig. 4.98 Sherwood Number vs. Schmidt Number for a Given Re in Presence of Acetic Acid in CO2 Saturated, 3.5 wt% NaCl Solutions at pH 5 & HAc Acid Conc. 3000 ppm.

5.52

log ShatRe=30000 = 4.312+0.9311*x Re = 30000; logSh = 4.3120 + 0.9311 x logSc; R = 1.00 logShatRe=50000 = 5.1837+0.204*x Re = 50000; logSh = 5.1837 + 0.2042 x logSc; R = 1.00 logShatRe=70000 = 3.6968+1.445*x Re = 70000; logSh = 3.6968 + 1.4450 x logSc; R = 0.99

5.50 5.48 5.46

Log Sh

5.44 5.42 5.40 5.38

Re = 30000 Re = 60000 Re = 90000

5.36 5.34 5.32 5.30 1.12

1.14

1.16

1.18

1.20

1.22

1.24

1.26

Log Sc

Fig. 4.99 Sherwood Number vs. Schmidt Number for a Given Re in Absence of Acetic Acid in CO2 Saturated, 3.5 wt% NaCl Solutions at pH 7.5.

204

Chapter Four


logShatRe=30000 = 4.4785+1.6218*x

6.58

Re = 30000; logSh = 4.4785 + 1.6218 x logSc; R =0.99 logShatRe=50000 = 5.4193+0.7836*x Re = 50000; logSh = 5.4193 + 0.7836 x logSc; R =1.00 logShatRe=70000 = 4.4673+1.6776*x Re = 70000; logSh = 4.4673 + 1.6776 x logSc; R =0.99

6.56 6.54 6.52 6.50

Log Sh

6.48 6.46

Re = 30000 Re = 60000 Re = 90000

6.44 6.42 6.40 6.38 6.36 6.34 6.32 6.30 6.28 1.12

1.14

1.16

1.18

1.20

1.22

1.24

1.26

Log S c

Fig. 4.100 Sherwood Number vs. Schmidt Number for a Given Re in Absence of Acetic Acid in CO2 Saturated, 3.5 wt% NaCl Solutions at pH 8.5. Table 4.25 The Physical Properties at Different Temperatures Uhlig, [2011] and Naftz et al., [2011] and David, [2007]. T (°C)

ρ (gm.cm-3)

µ (gm.cm-1.sec-1)

40 50 60 65 70 75

0.9923 0.9881 0.9833 0.9803 0.9773 0.9737

0.0066 0.0055 0.0047 0.0043 0.0041 0.0038

D (cm2.sec-1) x 104 1.5161 1.8660 2.2570 2.4976 2.6582 2.9099

Table 4.26 The pH and Hydrogen Ions Concentration of Solution Seo et al., [2012]. pH 3 5 7.5 8.5

[H+] (mol.cm-3) 1 x 10-6 1 x 10-8 3.16 x 10-11 3.16 x 10-12

205

Chapter Four


The Figures reveal straight lines in the plots of log Sh vs. log Sc. The dependence on Sc as shown by the index values (y in Equation 4.27) generally varies according to presence and absence of acetic acid (absence and presence of the protective film formation) and to the solution pH with positive index at pH’s 3, 5, 7.5 and 8.5. The slopes (△log Sh / △log Sc) for a given Re in presence and absence of acetic acid at different temperatures and pH values are shown in Table 4.27. Table 4.27 (△log Sh / △log Sc)Re in Presence and Absence of Acetic Acid at Different Temperatures and pH Values. Temp. (°C) 40 50 60 40 50 60 65 70 75 65 70 75

pH

HAc Conc. (ppm)

3

1000

5

3000

7.5

Blank Solution

8.5

Blank Solution

Re

△log Sh / △log Sc

30000 60000 90000 30000 60000 90000 30000 60000 90000 30000 60000 90000

0.684 1.419 1.142 0.119 1.416 1.415 0.931 0.204 1.445 1.622 0.784 1.678

The Sh, Re and Sc numerical values with presence and absence of acetic acid at different temperatures and a given pH values were computer analyzed using Levenberg- Marquardt estimation method regression analysis with the help of STATISTICA Computer Program v.10, in order to find the values of the constants c, x and y that give the best fitting of Equation (4.27). The results are shown in Table 4.28.

206

Chapter Four


Table 4.28 Experimental Results of c, x and y in Equation (4.27). Temp. (°C) 40 50 60 40 50 60 65 70 75 65 70 75

pH

HAc Conc. (ppm)

3

1000

5

3000

7.5

Blank Solution

8.5

Blank Solution

Speed of Rotation (rpm) 1000 1250 1500 1000 1250 1500 1000 1250 1500 1000 1250 1500

c

x

y

Correlation Coefficient

0.018

0.959

1.191

0.96

0.038

0.682

1.224

0.96

0.019

0.474

0.854

0.95

0.015

0.597

1.401

0.94

The functional form of the mass transport properties in solutions of pH = 3 in presence of acetic acid (absence of protective film formation) is shown by Equation (4.35): (

)

with coefficient of correlation = 0.96 The functional form of the mass transport properties in solutions of pH = 5 in presence of acetic acid (absence of protective film formation) is shown by Equation (4.36): (

)

with coefficient of correlation = 0.96 The functional form of the mass transport properties in solutions of pH = 7.5 in absence of acetic acid (presence of protective film formation) is shown by Equation (4.37): ( with coefficient of correlation = 0.95

207

)

Chapter Four


The functional form of the mass transport properties in solutions of pH = 8.5 in absence of acetic acid (presence of protective film formation) is shown by Equation (4.38): (

)

with coefficient of correlation = 0.94 The experimental Sh vs. calculated Sh is shown in Figures 4.101 through 4.104.

= 9.4196+0.8534*x Sh = 9.4196Sh + cal 0.8534 x (0.018 Re

120

0.959

Sc 1.191 ); R = 0.97

100

Sh

80

60

40

20

0 0

20

40

60

80

100

120

0.018 Re0.959 Sc 1.191

Fig. 4.101 Experimental vs. Calculated Sh Numbers in Presence of Acetic Acid at pH 3 & 1000 ppm HAc Acid.

208

Chapter Four

Results & Discussion Sh cal = 521.7777+0.9144*x 0.682

10000

Sh = 521.7777 + 0.9144 x (0.038 Re

Sc 1.225); R = 0.99

9000 8000

Sh

7000 6000 5000 4000 3000 2000 2000

3000

4000

5000

6000

7000

8000

9000

10000

0.038 Re0.682 Sc 1.224

Fig. 4.102 Experimental vs. Calculated Sh Numbers in Presence of Acetic Acid at pH 5 & 3000 ppm HAc Acid.

3E5

Sh cal = 77834.8114+0.7063*x Sh = 77834.8114+ 0.7063 x (0.019 Re

0.474

Sc 0.854 ); R = 0.97

2.9E5

2.8E5

Sh

2.7E5

2.6E5

2.5E5

2.4E5

2.3E5 2E5

2.2E5

2.4E5

2.6E5

2.8E5

3E5

3.2E5

0.019 Re0.474 Sc 0.854

Fig. 4.103 Experimental vs. Calculated Sh Numbers in Absence of Acetic Acid at pH 7.5.

209

Chapter Four

Results & Discussion Sh cal = 6.9235E5+0.706*x

Sh = 692350+ 0.706 x (0.015 Re 0.597 Sc1.401); R = 0.96

3.4E6 3.2E6 3E6

Sh

2.8E6 2.6E6 2.4E6 2.2E6 2E6 1.8E6 1.8E6

2E6

2.2E6

2.4E6

2.6E6

2.8E6

0.015 Re

0.597

3E6

Sc

3.2E6

3.4E6

3.6E6

3.8E6

1.401

Fig. 4.104 Experimental vs. Calculated Sh Numbers in Absence of Acetic Acid at pH 8.5. Equations (4.35 through 4.38) should be used with caution because of the scattering in the experimental results. The reasons for this scattering can be summarized as in the following: 1. The effect of temperature in changing the physical properties of the solution and especially the solubility of hydrogen. 2. During corrosion dissolution of the metal can lead to surface roughness and changes which can not be taken into consideration. 3. The presence of a hydrogen film on the surface interfering with the access of dissolved species and corrosion products accumulation to the surface. 4. The coverage of the metal surface with the corrosion products giving an additional barrier to the hydrogen diffusion. It is to be mentioned that in the present investigation the highest value of Sc (= 43.875) is at 40 °C which is rather low compared with previous investigations in the field of mass transfer as shown in Table 4.29. 210

Chapter Four


Table 4.29 Values of Sc and Temperature Used in Some Mass Transport Investigations. Ref. Hameed et al., [2009] Nesic et al., [1995] Bahar et al., [2013] Mahinpey, [2001] Wang and Nesic, [2003]

Sc 719 1000 572-949 1310-5655 562.8

T (°C) 15 20-80 30 0-25 25

Also most of these studies were performed at approximately room temperature or even lower. It was stated Wang and Nesic, [2003] that the measurements by the electrochemical techniques are for conditions of large Schmidt numbers where the region in which there is a measurable concentration gradient is thinner than the viscous layer (laminar sublayer) as shown in Figure 4.105. These points may explain at least, in part the different indices obtained in the present results.

Fig. 4.105 Sketch of the Computational Domain and A Typical Concentration Profile for A Dissolved Species Nesic et al., [2004]. Experimental studies on the corrosion of metals in deaerated acidic, neutral and alkaline solutions in presence and absence of acetic acid of different temperatures, pH values and speeds of rotation are not found in literature as they are normally performed at constant temperatures for varying Re values. 211

Chapter Four


Martin and Mokhtar, [2009] studied the corrosion of mild steel at 25°C in CO2 saturated and deaerated, 3% NaCl solution in presence of 60 ppm HAc and pH 5 at Re = 7500 - 45000. They found that the limiting current of hydrogen was in the range 550-1000 µA.cm-2. Bahar et al., [2013] studied the corrosion of carbon steel at 30°C in 0.1 N NaCl solution at Re = 13169-52676. They found that the limiting current of hydrogen was in the range 440-530 μA/cm2. Andijani, [2002] investigated the effect of rotational velocity on the initial corrosion behavior of a rotating carbon cylinder in deaerated of 1.0 M NaCl solution at 50°C, pH’s (4,6 and 9) and velocity (0,500,1000,2000 and 3000 rpm). As the velocity is increased further (3000rpm) the corrosion rate and corrosion potential do not change for all pH values. Hernandez et al., [2012] investigated the effect of hydrogen concentration and rotational velocity on the corrosion of rotating API X 70 steel cylinder at 20 °C and pH 3.9 at 1 atmo. In deaerated of 3.5% NaCl brine saturated with CO2. The limiting diffusion current of hydrogen varied from 225 to 675 µA. cm-2 with r.p.m increasing from 100 to 5000. The temperature increases the rate of reaction exponentially, e.g. hydrogen evolution reduction, while the diffusivity is a linear function of temperature, e.g. hydrogen reduction reaction. The temperature also has a reverse effect when corrosion is controlled by diffusion of hydrogen as it decreases its solubility. Palmer and Eldik, [1983] pointed out that in an open vessel saturated with carbon dioxide, allowing dissolved hydrogen to escape, the rate increases, the rate increases with temperature to about 80 °C and then fall to a very low value at the boiling point. The falling off of corrosion above 80 °C is related to a marked decrease of carbon dioxide and hydrogen solubility in water as the temperature is raised. 212

Chapter Four


Since the anodic and cathodic polarization curves change with temperature, the corrosion potentials may be decreased depending on the change in the partial polarization curves. Table 4.15 shows that most corrosion potentials were shifted to more negative values with increasing temperature in presence and absence of acetic acid for a given pH value and speed of rotation. In most mass-transfer studies in simulated flowing solutions the effect of temperature is to change the physical properties, which cause a change in Sc, a main dimensionless group in the mass-transfer correlations. The effect of Sc on Sh, plotted on log-log scale in presence and absence of acetic acid, at constant pH value and a given Re are shown in Figures 4.97 through 4.100. (△log Sh / △log Sc)Re, which represents the power y in Equation (4.27), ranged from 0.1 to 1.6 with most of the results at about 1.4 in the pH range 3-8.5. Many investigators agree with the Shaw and Hanratty, [1977] analogy for turbulent mass transfer conditions in which Sh changes with Sc7/15, although some slightly different powers for Sc were also reported. Most investigations in mass transfer have been performed in solutions of large Sc in order to be sure that the region in which there is a measurable concentration gradient is thinner than the viscous sublayer with one mass-transfer controlled reduction reaction at constant temperature, normally room temperature, as shown in Table 4.29. In some tests Mahinpey, [2001] the temperature was changed within a low region (0-25 °C), but the bulk concentration of the reduced ion remained the same. The present experiments in absence and presence of protective film were carried at low Sc (13.416-43.876) according to the test temperature of 40 to 75°C, in which the bulk concentration of dissolved hydrogen decreased with increasing pH, in addition to the other physical properties 213

Chapter Four


decreased with increasing temperature. Tables 4.25 and 4.26 show the effect of temperature and pH on these properties, respectively, in which the diffusivity is seen to increase with temperature, while the others (ρ, µ, [H+]) decreased at different rates, as the viscosity varies exponentially with temperature Perry and Green, [2000], the density slightly affected linearly by temperature Perry and Green, [2000]. The hydrogen concentration and diffusivity varied linearly with temperature George and Nesic, [2004]. Such variation explain the difference effects of temperature on the properties from which Sc is calculated, also the variation of hydrogen solubility affects Sh which results in an increasing straight lines relationship of data at different slopes as shown in Table 4.27. At pH’s (3 and 5) the effect of cementite (Fe3C) and siderite (FeCO3) film, which represents an additional resistance to the transfer of hydrogen to metal surface, makes Sh slightly affected by Sc. Also at pH’s (7.5 and 8.5) the effect of siderite (FeCO3) film, which represents an additional resistance to the transfer of hydrogen to metal surface, makes Sh accuracy affected by Sc. The type and thickness of these films, which have been studied in this investigation, at the optimum conditions in absence and presence of the protective film formation in weight loss technique may explain the effect of Sc on Sh. At 75 °C (Sc =13.416) the solubility of the film may be more than at 40 °C (Sc = 43.876), therefore the mass-transfer rate at Sc =13.416 can be higher than that at Sc = 43.876, therefore a different slope of (△log Sh / △log Sc)Re was obtained.

214

Chapter Four


4.3 Characterization of the Corroded Surface: 4.3.1 API X65 Mild Steel Protective Film Thickness Test: The protective film thickness is found from optical microscopy image of the cross-section as described in Figure 4.106 and porosity calculations of the protective film obtained are described in Appendix J.

4.3.2 API X65 Mild Steel Roughness Test: The arithmetic average values of the roughness are listed in Table 4.30: Table 4.30 Roughness Values for Corroded Surface of API X65 Mild Steel Specimens in CO2 –Saturated, 3.5 wt% NaCl Solution in Presence and Absence of Acetic Acid at the Optimum Conditions and Non Corroded Surface Specimens. API X65 Mild Steel (Non Corroded Surface) µm 0.025 0.021 0.028 Average 0.025

(Corroded Surface of API X65 Mild Steel in Presence of Acetic Acid) µm 0.292 0.287 0.288 0.289

(Corroded Surface of API X65 Mild Steel in Absence of Acetic Acid) µm 2.274 1.741 1.918 1.978

During corrosion dissolution of the metal can lead to surface roughness and changes which can be taken into consideration. The roughness of corroded surface in absence of acetic acid is greater than that in presence of acetic acid because during the corrosion process, occur corrosion product formation on the corroded steel surface and a new phases is formed of a porous cementite (Fe3C) layer and siderite FeCO3 phase, while in absence of acetic acid a new phase is formed, i.e. siderite FeCO3 phase. The source of the Fe3C is probably the steel itself Nor et al., [2011], because, under the experimental conditions used, the formation of Fe3C from iron and carbon dioxide is thermodynamically allowed Nor et al., [2011]. The weight and dimensions of these phases 215

Chapter Four


are different from that of base metal that is already replaced by the non protective and protective film formation. Eventually, the differences in roughness values measured in perpendicular directions are not the same, which indicates the shape of asperity in the two directions is not the same. Figure 4.107 a & b & c shows the surface morphology and microstructure of non corroded surface and corroded surface for corroded specimens of API X65 mild steel in CO2-saturated, 3.5 wt % NaCl solution after 3h at the optimum conditions in presence of acetic acid (presence of non protective corrosion layer formation) and in absence of acetic acid (presence of protective corrosion layer formation) respectively. In Figure 4.107 c, it is clearly noticed that the rough surface in absence of acetic acid is higher than in presence of acetic acid due to the protective and non protective corrosion layers formation created on the surface. The types of this layer appear as spherical or semispherical. Precipitation process depends on surface cleaning and the presence of a small amount of impurities (especially C) slows down the precipitation rate and determines the scale growth and its protectiveness of FeCO3 and Fe3C in these sites. These observations are also shown in Figure 4.107 b. Table 4.30, also shows the roughness measurement values for non corroded surface and corroded surface of API X65 mild steel specimens in CO2-saturated, NaCl environment in presence and absence of acetic acid at the optimum conditions. Martin and Mokhtar, [2009] proved that the roughness of corroded surface for BS-970 mild steel in absence of acetic acid and sodium acetate anhydrous in CO2 saturated, 3% NaCl aqueous in turbulent flow conditions is greater than the roughness in presence of acetic acid and sodium acetate anhydrous.

216

Chapter Four


Resin

Metal L = 107.307 µm

Film

Epoxy Resin

Fig. 4.106 SEM and Optical Microscopy of the Cross Sections of Corroded API X65 Mild Steel Specimen in CO2-Saturated , 3.5 wt% NaCl Solution at the Optimum Conditions Showing the Thickness of the FeCO3 Layer Formed.

217

Chapter Four


(a)

218

Chapter Four


(b)

219

Chapter Four


(c) Fig. 4.107 Surface Morphology and Microstructure of Corroded API X65 Mild Steel Specimens in CO2-Saturated, 3.5 wt% NaCl Solution at the Optimum Conditions: (a) Non Corroded Surface (b) Presence of Acetic Acid (Absence of Protective Film Formation) (c) Absence of Acetic Acid (Presence of Protective Film Formation).

220

Chapter Four


4.3.3 API X65 Mild Steel X-Ray Diffraction: X-ray diffraction (XRD) analyses results of the corroded surface (i.e., the corrosion product formed from the corroded surface) of API X65 mild steel specimens tested after exposure period 3 hour at the optimum conditions (obtained in weight loss technique ) in CO2 saturated, 3.5 wt% NaCl solutions in presence and absence of acetic acid are shown in

90 100

Figures 4.108 a & b.

50 60 20 30 40

Intensity

70

80

Fe3C

0

10

FeCO3

2

angle / degrees

(a)

FeCO3

FeCO3

FeCO3 FeCO3

2

angle / degrees

(b) Fig 4.108 XRD Pattern of Corroded API X65 Mild Steel Surface Specimens in CO2-Saturated, 3.5 wt% NaCl Solution at the Optimum Conditions in: (a) Presence of Acetic Acid (Absence of Protective Film Formation) (b) Absence of Acetic Acid (Presence of Protective Film Formation). .

221

Chapter Four


In Figure 4.108 a, clearly revealed that there are new phases obtained on the surface of specimens in presence of acetic acid (presence of the non protective film formation) show that the scale consist of cementite Fe3C and siderite (FeCO3) phase while, phases obtained in absence of acetic acid (presence of the protective film formation) show that the scale consist of FeCO3 phase only, as shown in Figure 4.108 b. X-ray analysis data are described in Tables J.2 and J.3 in Appendix J. . A peak near 30° 2 is consistent with siderite (FeCO3). FeCO3 is the most important film that formed at carbon steel in sweet/CO 2 environment. This film can be function as a protective film for the carbon steel from the CO2 corrosion. It is believed that the corrosion rate was decreased when this film formed as indicated shown in Tables 4.2 and 4.7. The film formation is strongly depending on the thermodynamics and kinetics of FeCO3 precipitation. In principle, the precipitation process comprises two steps, nucleation and particle growth. With increasing pH, making the rate decreased solubility of FeCO3 resulting high deposition rate Fang, [2006].The morphology of the film therefore depends on the dominating steps Kermani and Morshed, [2003] and Daugstad, [1998]. Protective film formation is accelerated by measures that restrict the transport of reaction products from the surface Daugstad et al., [2000]. This comes in good agreement with Forero et al., [2013] as he has studied the X-ray analysis of the corrosion scales formed on API 5L X70 and X80 steel pipe in deareated 1% NaCl solution containing CO2 and found the protective layer of (FeCO3) film formed on the surface of both protective steels in the range 60 to 80 °C and observed the porous and discontinuous Fe3C (from the steel microstructure) layer formed in the range of 40 to 60 °C.

222

Chapter Four


4.3.4 API X65 Mild Steel Hardness Test: The Vickers micro-hardness testing (VMH) results are listed in Table 4.31. Table 4.31 Micro-Hardness Values for Corroded API X65 Mild Steel Surface Specimens in CO2 –Saturated, 3.5 wt% NaCl Solution in Presence and Absence of Acetic Acid at the Optimum Conditions and Non Corroded Surface Specimens. API X65 Mild Steel (Non Corroded Surface) Micro-Hardness Value VMH 135 134 138 136 137 Average 136

(Corroded Surface of API X65 Mild Steel in Presence of Acetic Acid) Micro-Hardness Value VMH 184.8 180.0 182.2 181.5 183.0 182.3

(Corroded Surface of API X65 Mild Steel in Absence of Acetic Acid) Micro-Hardness Value VMH 307.8 269.6 291.7 280.9 287.7 287.5

From above results, it is obvious that the hardness of corroded surface of API X65 mild steel in absence of acetic acid is higher than in presence it due to the presence of chemical compounds (Fe3C and FeCO3) formed of the corroded steel surface, these compounds (FeCO3) increase the mechanical properties of the mild steel and increase the corrosion resistance and for less extent of compounds (Fe3C). Table 4.31 also shows that the hardness of corroded surfaces in absence and presence of acetic acid is higher than that of uncorroded specimens. This refers to the great influence of the existence of iron carbide and ferrous carbonate on the hardness values in presence and absence of acetic acid respectively. Crolet et al., [2004] illustrate that the hardness of protected API X52, X60, X65 and X70 mild steel alloys are higher than unprotected

223

Chapter Four


specimens by precipitation process of an insoluble corrosion product FeCO3 on CO2 corrosion in turbulent flow conditions. This comes in good agreement with Ogundele and White, [1987] as he has studied the hardness of protected API-L80 steel and found greater hardness in comparison with the unprotected alloy specimens from corrosion of CO2 saturated, brine solution using precipitated iron carbonate with and without influence of dissolved hydrocarbon gases and variable water chemistries.

224

Chapter Five Conclusions & Recommendations for Further Work

Chapter Five

Conclusions & Recommendations for Further Work

5.1 Conclusions:

T

he following conclusions might be withdrawn from this investigation:

5.1.1 Weight Loss Technique: 1. The second order polynomial regression analysis of the objective function (corrosion rate) describe the behavior of the process in both absence and presence of the protective film formation with mean absolute percentage error 1.2 % and 0.21 % respectively in terms of temperature, solution pH, HAc acid concentration and speed of rotation. 2. The corrosion rate of API X65 mild steel in CO2 saturated, 3.5 wt % NaCl solution in presence and absence of acetic acid, increases with increasing temperature, acetic acid concentration and speed of rotation, and decreased with increasing solution pH. 3. The variation of corrosion rate of API X65 mild steel in CO 2 saturated, 3.5 wt % NaCl solution with temperature obeys the Arrhenius Equation in presence and absence of acetic acid (absence and presence of the protective film formation). 4. The values of activation energy Ea from Arrhenius Equation in absence of acetic acid (presence of the protective film formation) are significantly influenced by temperature, solution pH and speed of rotation which is high at 1000 rpm indicating that the reaction needs more energy to occur, lower at 1500 rpm indicating improvement of protective film formation efficiency with temperature and solution pH increasing. 5. It is found that the values of enthalpy of activation (ΔH*) from transition state Equation were positive, indicating that the corrosion process is endothermic reaction. The value of entropy of activation 225

Chapter Five

S  


were negative in absence and presence of the protective film

formation which indicate that the transition state of the rate determining recombination step for hydrogen evolution reaction represents a more orderly arrangement relative to initial state. 6. Values of free Gibbs energy change (△G) were higher in presence than in absence of the protective film formation at temperature range (65-75 °C), pH 8.5 and 1000 rpm than the values of other conditions used, which indicate the strong binding of formation the protective film to the metal surface as compared with formation at other conditions used. 7. The optimum conditions as predicted from Equations (4.4) and (4.11) are 45.4 °C, pH 4.8, 2178.5 ppm HAc and 1296.6 rpm in absence of the protective film formation while 68.7 °C, pH 7.9 and 1425.8 rpm in presence of the protective film formation. 8. The weight loss technique measurement evaluation of corrosion API X65 mild steel in CO2 saturated, 3.5 wt % NaCl solution in presence and absence of acetic acid (absence and presence of the protective film formation) were obtained at the optimum conditions shows a super suppression of corrosion rate in following order: FeCO3 film > in absence of HAc

(Fe3C + FeCO3) film in presence of HAc

9. The analysis of statistical Full Factorial Experimental Design (FFED), generally, shows that the square and interaction effects on the corrosion rate by weight loss technique (within the studied range) is less pronounced compared with the main variables except the high interaction effect of (T x pH) in presence of HAc. 10. The effect of the temperature, solution pH, HAc acid concentration and speed of rotation on the corrosion rate in absence and presence of the protective film formation is in the following order: 226

Chapter Five


In absence of the protective film formation: Temperature > pH > HAc Acid Conc. > Speed of Rotation In presence of the protective film formation: Temperature > pH > Speed of Rotation 11.The primary corrosion product of API X65 mild steel is ferrous carbonate (FeCO3) at high temperatures, high pH’s (alkaline media) and absence of acetic acid, which could act as a protective film so that CO2 corrosion rate can be reduced. Based on the experimental results, corrosion rate of API X65 mild steel in CO2 saturated, 3.5 wt % NaCl solution varied from 12.44 to 37.09 gmd. But in presence of acetic acid, corrosion rate varied from 40.06 to 130.46 gmd. This is categorized as fairly high which could endanger flowline and pipeline in oil and gas production facilities where wall thinning process is severely occurred which tends to cause pipe leaks or failure. Therefore flowline and pipeline which subject to CO2 corrosion need to be protected during in service.

5.1.2 Potentiodynamic Polarisation Technique: In this investigation, electrochemical measurments for API X65 mild steel were studied under the effet of different conditions in absence and presence of the protective film formation. It is observed that: 1. The corrosion current densities in presence and absence of acetic acid increase with increasing temperature and speed of rotation and decrease with increasing pH value in absence of acetic acid. The same behavior was observed from the weight loss technique. 2. The protective film formation at pH 7.5 inhibts the corrosion of API X65 mild steel in CO2 saturated, 3.5 wt % NaCl solution in absence of acetic acid by affecting both anodic and cathodic partial reaction with more effect on the anodic one. The effect at pH 8.5 on anodic area is very large as compared with its effect on cathodic area. Under 227

Chapter Five


no protective film formation at pH’s (3 and 5) show a mixed effect with predominant on the cathodic reaction. 3. The values of corrosion potential showed that under no and in the protective film formation shift to the active direction with increasing both temperature & speed of rotation. 4. The values of Tafel slopes bc & ba in absence and presence of the protective film formation were approximately the same. However, the slightly decreasing in values of bc and ba may be attributed to the formation of the protective film. 5. The value  , (i.e.,

icorr ) was between 0 and 1, therefore,  –Model ilim

used in present work suggest that the mechanism of API X65 mild steel corrosion in CO2 saturated, 3.5 wt % NaCl solution was a mixed control (i.e., both mass transfer and charge transfer will control the corrosion process). In absence of the protective film formation at pH 3 high values of  were obtained, this indicate the diffusion control of the corrosion process. 6. The values of polarization resistance (Rp) represent the charge transfer resistance resulting from the separation the charges across the solid interface to the outside edge of the double layer, these values were decreased with increasing of temperature and speed of rotation. The values of Rp were higher in presence of protective film formation at pH 8.5 as compare with presence of protective film at pH 7.5. 7. Tafel extrapolation method and McLaughlin method can be applied to determine the corrosion kinetic parameters with high accuracy. Tafel extrapolation method is better than McLaughlin method in activation control systems. While McLaughlin method can be applied successfully with system contains mass transfer effects.

228

Chapter Five


8. Electrochemical measurements generally, have shown that the presence of acetic acid (HAc) strongly affects the limiting current, which is mass transfer controlled. The anodic reaction (iron dissolution) was unaffected with increasing HAc concentration at high pH. 9. As the temperature increases, the HAc effect is much more pronounced, leading to higher corrosion rates. The HAc corrosion becomes mass-transfer controlled at elevated temperatures, making it more prone to flow effects. 10. The β-Model shows good approach for calculating the corrosion kinetics parameters in CO2 saturated, 3.5 wt % NaCl solutions in absence and presence of protective film formation when the concentration polarization is effective. 11. Temperature, solution pH value and speed of rotation, as well as HAc acid concentration may change the corrosion regime from activation in to mixed or mass transport control. 12. The values of mass transfer correction factor (λ) represent the effect of mass-transport in charge transfer process, these values depend on overpotential η, β values and cathodic Tafel slope bc, these values will approach unity at low overpotential and it decreases as overpotential increases. Generally in absence of acetic acid, the values of (λ) are closest to each other and almost unite value at pH’s (7.5 and 8.5) as compare with presence of acetic acid at pH’s (3 and 5) at different temperatures and speeds of rotation due to the protective film formation as diffusion barrier is accelerated by measures that restrict the transport of reaction products from the surface.

5.1.3 Characterization of the Corroded Surface Techniques: 1. The roughness and hardness for non-protective layer (i.e., Fe3C & FeCO3) in presence of acetic acid (HAc) are found to be lower than 229

Chapter Five


protective film formation (i.e., FeCO3 only) in absence of acetic acid (HAc). While the porosity of non protective layer in presence of acetic acid is higher compared than the protective layer in absence of acetic acid. 2. As seen from scanning electron microscopy (SEM) and optical microscopy (CMOMT) pictures in Figure (4.107 a, b &c), there is uniform corrosion on the metal and no evidence of localized attack is observed.

5.2 Recommendation for Further Work: 1. The mild steel inhibition in deareated and oxygen saturated acid solution at different temperatures and Reynolds number values in fully developed velocity and mass transfer conditions can be suggested for further investigation. 2. Further work should be performed to investigate the influence of some drag reducing agents on the corrosion inhibition of mild steel through the determination of mass transfer coefficient. 3. An investigation using Electrochemical Impedance Spectroscopy (EIS) technique is requested, to analyze the results, and for better understanding the mechanism of anodic and cathodic reactions in the corrosion process. 4. Constructing a Pourbaix Diagram and validated with the experimental data over the studied temperature range, on the assumption that Fe +2, solid FeCO3 and solid Fe3C are the main corrosion products to be expected from CO2 corrosion of mild steel in the pH range of (3-8.5). 5. The electrochemical kinetics of CO2 corrosion at temperatures higher than 75 °C.

230

References

References

References

 Abd El-Lateef, H.M., Abbasov, V. M., Aliyeva, L.I. and Ismayilov, T.A.,“Corrosion Protection of Steel Pipelines Against CO2 CorrosionA Review”, Chemistry, Vol.2, Issue 2, pp. 52-63, (2012).  Agotnes, N.O. and Hemmingsen, T. et al., “Comparison of Corrosion Measurements by Use of AC-impedance, LPR and Polarization Methods on Carbon Steel in CO2 purged NaCl Electrolytes”, Corrosion, No.27, NACE, (1999).  Amri, J., “On Growth and Stifling of Localized Corrosion Attacks in CO2 and Acetic Acid Environments: Application to the Top-of-Line Corrosion of Wet Gas Pipelines Operated in Stratified Flow Regime”, Ph.D. Thesis, French Institut Polytechnique de Grenoble, France, (2009).  Anderko, A., McKenzie, P. and Young, R. D., “Computation of Rates of General Corrosion Using Electrochemical and Thermodynamic Models”, Corrosion, Vol. 57, No.3, (2001).  Andijani, I., “Ac Impedance Studies on the Corrosion Behavior of Carbon Steel in Deaerated Chloride Solution”, Corrosion, No.12 (2002).  Andijani, I. and Turgoose, S., “Studies on Corrosion of Carbon Steel in Deaerated Saline Solutions”, (1999).  Aprael, A.S., Khadom, A. A. and Ibraheem, H.F.,“Peach Juice as an Anti-Corrosion Inhibitor of Mild Steel”, Anti-Corrosion Methods and Materials, Vol. 58, No. 3, pp.116-124, (2011).  Asmara, Y. P. and Mokhtar, C.I., “Study corrosion prediction models in CO2 environment”, ICPER, KL, (2008).  Bagotsky, V.S., “Fundamentals of Electrochemistry”, 2nd Edition, John Wiley & Sons, Inc., (2006).

231

References

References

 Bahar,S.S., Hussein, W.N. and Alwan, H.H., “ Corrosion Behavior of Carbon Steel Under Flowing Condition”, Babylon Journal of Engineering Sciences, (2013).  Bala, H. B., Br. Corros. Jr., Vol. 23, No. 29, (1988).  Bin Muhammad, M.F., Master Thesis, University Technology Petronas, (2013).  Bockris, J.O.M. and Reddy, A.N., “Modern Electrochemistry”, Vol.2, Plenum Press, New York (1977).  Box, H. and Hunto, “Statistics for Experiments”, 2nd Edition, JohnWiley and Sons. (2005).  Buchanan, R. A. and Stansbury, E. E., “Fundamental of Electrochemical Corrosion”, 1st Edition, ASM International (2000).  Burke, P., “Synopsis: Recent Progress in the Understanding of CO2 Corrosion”, Corrosion, No. 2, NACE, (1984).  Chen, H.J., Hong, T. and Jepson, W.P., “High Temperature Corrosion inhibition Performance of Imidazoline and Amido”, No. 35, NACE International Conference, (2000).  Cottis, R. A., “Electrochemical kinetics and thermodynamics”. (2002).  Crolet et al., Technical performance “microhardness”, (http://www. Metalast.html) version 2.0, metalast Tech. Center, (2004).  Crolet, J.L. and Bonis, M. R., “The Role of Acetic Acid in CO2 Corrosion of Carbon Steel”, Corrosion, No. 5272, NACE, (2005).  Crolet, J.L., Thevenot, N. and Nesic, S., “Role of Conductive Corrosion Products in the Protectiveness of Corrosion Layers”, Corrosion Science, NACE International, (1998).  David, R. L., “Handbook of Chemistry and Physics”, 87th Edition, CRC Press Inc., (2007). 232

References

References

 David, T. and James, T., “Corrosion Science and Technology”, CRC Press LLC, (1998).  Devore, P., “The Exploration and Analysis Data”, 5th Edition, Thomson Learning, Belmont, USA, (2005).  De Waard, C., Lotz, U. and Dugstad, A., “Influence of Liquid Flow Velocity on CO2 Corrosion:A Semi-Empirical Model”, Corrosion, No. 128, NACE, Houston, Texas (1995).  Dougherty, J. A., Annual Corrosion Conference, No. 4376, NACE, (2004).  Driver, R. and Meakins, R.J., Br. Corros. Jr., Vol.9, No.4, p.227, (1974).  Dugstad, A., “Fundamental Aspects of CO2 Metal Loss Corrosion Part 1: Mechanism”, Corrosion, No. 6111, NACE, (2006).  Dugstad, A., Hammer, H. and Seirstein, M., “Effect of Steel Microstructure upon Corrosion Rate and Protective Iron Carbonate Film Formation”, Corrosion, No. 23, NACE, Houston, Texas (2000).  Dugstad, A., “Mechanism of Protective Film Formation during CO 2 Corrosion of Carbon Steel,” Corrosion, Vol. 98, No. 31, NACE, Houston, Texas, (1998).  Dugstad, A., “The Importance of FeCO3 Supersaturation of Carbon Steel”, Corrosion, Vol.92, No. 14, NACE, (2004).  Egil, G. and Katerina, B., “Solution Chemistry Effects on Corrosion of Carbon Steels in Presence of CO2 and Acetic Acid”, Corrosion, No. 6364, NACE, (2006).  Einar, B., “Corrosion and Protection”, Springer, London, (2004).  Fajardo, V., Brown, B., Young, D. and Nesic, S., “Multipronged Characterization of the Corrosion Product Layer Formed on Mild Steel after Exposure to Acetic Acid”, Corrosion, No. 2453, NACE, (2013). 233

References

References

 Fang, H., M.Sc. Thesis, “Low Temperature and High Salt Concentration Effects on General CO2 Corrosion for Carbon Steel”, The Russ College of Engineering and Technology of Ohio University, Ohio, USA, (2006).  Farelas, F., Galicia, M., Brown, B., Nesic, S. and Castaneda, H., Corros. Sci., Vol. 52, No. 509, (2010).  Faysal, F. E., “Electrochemical Evaluation of API-X100 Pipeline Steel in Simulated Carbon Dioxide Corrosion Environments”, Master Thesis, University of British Columbia (2011).  Forero, A.B., Nunez, M.M.G. and Bott, I. S., “Analysis of the Corrosion Scales Formed on API 5L X70 and X80 Steel Pipe in the Presence of CO2”, Meterials, No. 8, (2013).  Forsal, M., Touhami, E., Mernari, B., El Hajri, J. and Filali Baba, M., “Use of Experimental Designs to Evaluate the Influence of 2Mercaptobenzimidazole on the corrosion of Mild Steel in HCl (1M) Environment in the Presence of Alcohol Ethoxylate”, Electrochim. Acta, Vol. 28, No. 3, (2010).  Frankel, G.S., “Localized Corrosion of Metals; A review of the Rate Controlling Factors in Initiation and Growth”, Fontana Corrosion Center, (www.mse-gsf1.eng.ohio-state.pdf.), (2004).  Gabe, D.R. and Makanjoula, P. A., EFCF Publication Series No. 15, “Electrochemical Engineering”, AIChE Symposium Series, (98), p. 309, (1986).  Garsany, Y., Pletcher, D. and Hedges, B., “The Role of Acetate in CO2 Corrosion of Carbon Steel”, Corrosion, No.2273, NACE, Denver, Texas (2002).  George, I., Omkar, N. and Nafday, A., “Film formation and CO2 corrosion in the presence of acetic acid”, Master theses of Fritz, j., 234

References

References

Dolores, and Russ, H., College of Engineering and Technology, Ohio University, (2004).  George, K. and Nesic, S., “Electrochemical Investigation and Modeling of Carbon Dioxide Corrosion of Carbon Steel in The Presence of Acetic Acid”, Corrosion, No.4379, NACE, (2004).  George, K.S. and Nesic, S., “Investigation of Carbon Dioxide Corrosion of Mild Steel in the Presence of Acetic Acid-Part 1: Basic Mechanisms”, Corrosion, Vol.63, No.2, (2007).  George, K.S., “Electrochemical Investigation of Carbon Dioxide Corrosion of Mild Steel in the Presence of Acetic Acid”, M.Sc Thesis, Ohio University, USA, (2003).  Gulbrandsen, E. and Bilkova, K., Corrosion, Vol.6, No. 06364, NACE, (2006).  Gulbrandsen, E., Nesic, S., Stangeland, A., Buchardt, T., Sundfaer, B., Hesjevik, S.M. and Skjerve, S., “Effect of precorrosion on the performance of inhibitors for CO2 corrosion of carbon steel”, Corrosion, No.13, (1998).  Gunaltum, Y.M. and Larrey, D., “Correlation of Cases of Top of Line Corrosion with Calculated Water Condensation Rates”, Corrosion, No. 71, NACE, Houston, Texas (2000).  Hameed, K.W., Khadom, A. A. and Rahman, A. M., “Mass Transfer Coefficient during Cathodic Protection of Low Carbon Steel in Seawater”, Int J. Al-Khwarizmi Engineering, Vol.5, No.1, pp. (2032), (2009).  Halvrsen, A.M.K. and Anderson, T.R., “pH stabilization for internal corrosion protection of pipeline carrying wet gas with CO2 and acetic acid”, Corrosion, Vol. 3, No. 3329, NACE, Houston, Texas (2003).

235

References

References

 Hedges, B. and McVeigh, L., “The Role of Acetate in CO2 Corrosion: The Double Whammy”, Corrosion, No. 12, NACE, Houston, Texas (1999).  Hernandez, J., Munoz, A. and Genesca, J., “Formation of ironcarbonate scale-layer and corrosion mechanism of API X70 pipeline steel in carbon dioxide-saturated 3% sodium chloride”, Corrosion, No.10, pp. 251-258, (2012).  Ibraheem, M.A., Fouda, A.E. ,Rashad, M.T. and Sabbahy, F.N., “Sweet Corrosion Inhibition on API 5L-B Pipeline Steel”, Corrosion, No.10, (2012).  James, A. D., “A Review of the effect of organic acids on CO2 corrosion”, Corrosion, No. 04376, NACE, Houston, Texas (2004).  Jeff Wu, C. F. and Michael, S.H., “Experiments: Planning, Analysis and Optimization”, 2nd Edition, John Wiley and Sons, Inc., New York, USA, (2009).  Jezmar, J., “Monitoring Methods of Cathodic Protection of Pipe lines”, Corros. Meas., No. 2, pp.1-13, (2002).  Jiabin, H., William, J.C. and Jinsou, Z., “Effect of Sodium Chloride on Corrosion of Mild Steel in CO2-Saturated Brines”, J. Appl. Elect., Vol. 41, No. 6, p.741,(2011).  Jiang, X., Zheng, Y. G., Qu, D. R. and Ke, W., Corr. Sci. Vol. 48, No. 3091, (2006).  John, N. and Karen, T., “Electrochemical Systems”, 3rd Edition, JohnWiley and Sons, Inc (2004).  Joosten, M.W., Kolts, J., Hembree, M. and Achour, M., “Organic Acid Corrosion in Oil and Gas Production”, Corrosion, No.2294, NACE, Denver, Texas (2002).

236

References

References

 Jun, H.K. and In, S.K., “Flow-Accelerated Corrosion Behavior of SA106 Gr.C Steel in Alkaline Solution Characterized by Rotating Cylinder Electrode”, Electrochem. Soc., Vol. 32, No.6, pp.595-604, December, (2000).  Kapusta, S.D., Pots, B.F.M. and Rippon, I.J., “The Application of Corrosion Prediction Models to the Design and Operation of Pipelines”, Corrosion, No. 633, NACE, Houston, Texas (2004).  Kermani, M.B. and Morshed, A., “Carbon dioxide corrosion in oil and gas production-A Compendium”, Corrosion, No. 59, pp. 659-683, (2003).  Kermani, M.B. and Morshed, A., Corrosion, No. 131, NACE, Houston, Texas (2003).  Khadom, A.A., “Application of Two Theoretical Models to Corrosion of Mild Steel and Copper-Nickel Alloy in Hydrochloric Acid”, Diyala Journal of Engineering Sciences, pp.406-413, (2010).  Khattree, R. and Rao, C.R., “Handbook of Statistics”, Vol. 22, Elsevier Science B.V, (2003).  Korobove, V.I. and Medvedeva, I.A., 51 ISE Meet., Warsaw, Poland, abstract 3/6, (2000).  Koteeswaran, M., “CO2 and H2S Corrosion in Oil Pipeline, p.79, University of Stavenger, (2010).  Kreyszig, E., “Advanced Engineering Mathematics”, John Wiley & Sons, 10th Edition, (2011).  Lazarevic, A., Marinkovic, V. and Lazarevic D., “Expanded nonlinear mathematical models in the theory of experimental design: a case study”, 10th International

Conference

″Research

and

Development in Mechanical Industry″-RaDMI, D. Milanovac, pp. 304311, (2010). 237

References

References

 Li, H., Ph.D. Thesis, University of Ohio (2011).  Li, J. and Zuo, J., “Influences of temperature and pH value on the corrosion behaviors of X80 pipeline steel in carbonate/bicarbonate buffer solution”, Chinese Journal of Chemistry, Vol. 26, pp. 17991805, (2008).  Mahinpey, N., “An Electrochemical Investigation of Mass Transfer in a Pipe and a Simplified Bifurcation Model”, Ph.D. Thesis, University of Toronto, (2001).  Mansfeld, F, Corrosion, Vol.61, No.11, p.1019, (2005).  Marcus, P. and Herbelin, J.M., Corrosion Science, Vol. 34, No. 1123, (1993).  Martin, C.F. and Mokhtar, C. I., Effect of Low Concentration Acetic Acid

on

CO2

Corrosion

in

Turbulent

Flow

Conditions”,

Corrosion,Vol.34, No. 12, pp.129-146, (2009).  McCaffrty, E., “Introduction to Corrosion Science”, Alexandria VA 22309, USA, (2010).  McLaughlin, B.D., “A new approach for determining corrosion currents and Tafel slopes”, No. 37, pp.723-726, (1981).  Migahed, M.A., “Corrosion inhibition of Steel Pipelines in Oil Fields by N, N-di (poly oxy ethylene) Amino Propyl Lauryl Amide”, Progress in Organic Coatings, Vol. 54, No.2, pp.91-98, (2005).  Mokhtar, I.C., “Prediction CO2 corrosion with the presence of acetic acid”, Ph.D. Thesis, UMIST, United Kingdom, (2005).  Montgomery, D.C., “Design and Analysis of Experiments”, 6th Edition, John Wiley and Sons, New York, USA, (2005).  Mysara, E. M., Noaman, E. and Mokhtar, I.C., “ A Computational Model for Simulation of CO2 Corrosion Rate along Pipelines”, 3rd

238

References

References

International Conference on Solid State Science & Technology, Malaysia, December 1-3, (2010)  Nafday, O. A., “Film Formation and CO2 Corrosion in the Presence of Acetic Acid”, M.Sc Thesis, Ohio University, USA, (2004).  Naftz, L., Millero, J., Jones, B. and Reed Green, W., “An Equation of Sate for Hypersaline Water in Great Salt Lake,Utah, USA”, No. 17, pp.(809-820),(2011).  Nagy, Z. and Thomas, D.A., J. Electochem. Soc., Vol. 133, p. 2013, (1986).  Nervana, A. A., “The Effect of Temperature and pH on the Corrosion Rate of Carbon Steel in 1 M NaCl”, (2010).  Nesic, S., “Key Issues Related to Modelling of Internal Corrosion of Oil and Gas Pipelines - A Review”, Corrosion Science, Vol.49, No.12, pp. 4308-4338, (2007).  Nesic, S. and Nafday, O.A., “Iron Carbonate Scale Formation and CO2 Corrosion in the Presence of Acetic Acid”, Corrosion, No. 5295, (2005).  Nesic, S., “Evaluating and Qualifying Oilfield and Refinery Corrosion Inhibitors in the laboratory”, ASTM G 170-01, NACE International, pp.6-8, (1996).  Nesic, S., Novdsveen, M., Nyborg, R. and Stangeland, A., “A Mechanistic Model for CO2 Corrosion with Protective Iron Carbonate Films”, Corrosion, No. 1040, NACE, Houston, Texas (2001).  Nesic, S., Nordsveen, M., Nyborg, R. and Stangeland, A., “A Mechanistic Model for Carbon Dioxide Corrosion of Mild Steel in the Presence of Protective Iron Carbonate Films-Part 2: A Numerical Experiment”, Corrosion, Vol.59, No.6, pp. 489-497, (2003).

239

References

References

 Nesic, S., Pots, B.M., Postlethwaite, J. and Thevenot, N., “Superposition of Diffusion and Chemical Reaction Controlled Limiting Currents-Application to CO2 Corrosion”, J.Corrosion Science and Engineering, Vol.1, No. 3, (1995).  Nesic, S., Solvi, G.T. and Enerhaug, J., “Comparison of the Rotating Cylinder and Pipe Flow Tests for Flow-Sensitive Carbon Dioxide Corrosion”, Corrosion, Vol.51, No. 10, p.773, (1995).  Nesic, S., Xiao, Y. and Pots, B.M., “A quasi 2-D Localized Corrosion Model”, Corrosion, No. 4628, p.15, NACE, (2004).  Nor, A.M., Suhor, M.F., Mohamed, M.F., Singer, M. and Nesic, S., “Corrosion of Carbon Steel in High CO2 Environment: Flow Effect”, Corrosion, No. 11242, pp.1-18, NACE, (2011).  Nyborg, R., “CO2 Corrosion Models for Oil and Gas Production Systems”, Corrosion, No. 10371, NACE, (2010).  Nyborg, R., Corrosion, No. 48, NACE, (1998).  Nyborg, R., Corrosion, No. 233, NACE, Houston, Texas (2002).  Nyborg, R. and Dugstad, A., Corrosion, No. 642, NACE, (2003).  Nyborg, R., “Internal Corrosion and Protection of it in Oil and Gas Pipeline”, (http: //www.touchoilandgas.com), (2004).  Ogundele, G. and White, W., “Observations on the influence of dissolved hydrocarbon gases and variable water chemistries on corrosion of an API-L80 steel”, Corrosion, Vol. 43, pp. 665-673, (1987).  Palmer, D. A. and Eldik, R. V., “The Chemistry of Metal Carbonato and Carbon Dioxide Complexes”, Chemical Reviews, Vol. 83, No. 6, pp. 651-731, (1983).

240

References

References

 Paolinelli, L., Perez, T. and Simiso, S., “The effect of pre-corrosion and steel microstructure on inhibitor performance in CO2 corrosion”, Corrosion Science, No.50, pp.2456–2464, (2008).  Perez, N., “Electrochemistry and Corrosion Science”, Academic Press, London, (2004).  Perry, R.H. and Green D.W., “Chemical Engineers Handbook”, 8th Edition, McGraw-Hill, United States, (2000).  Peter, A. and Julio, de P., “Elements of Physical Chemistry”, 5th Edition, Oxford University Press, (2009).  Philip, J. W., Robert, A. and Webster, R., “Turbulent diffusion”, School of Civil and Environmental Engineering, Georgia Institute of Technology, (2002).  Pourbaix, M., “Literatures on Electrochemical Corrosion”, p.49, Plenum, New York, (1973).  Rana, A. M., “Thermodynamic and kinetic study for the corrosion of aluminium”, Lambert Academic Publishing, (2013).  Roberge, P., “Handbook of Corrosion Engineering”, 2nd Edition, McGraw-Hill., Inc., (2012).  Roberge, R. P., “Electrochemical kinetics principles nearest diffusion layer”. (Internet site: www.corrosion_doctor.org), (2004).  Robert, G. K., Scully, J.R., Shoesmith, D. and Rudolph, G. B., “Electrochemical

Techniques

in

Corrosion

Science

and

Engineering”, 2nd Edition, Marcel Dekker, Inc., (2003).  Robert, B. and James, P.M., “Improving the Design of the Scientists and Engineers Statistical Data System (SESTAT)”, 1st Edition, Washington, D.C., (2003).  Robertson, J. and Forrest, J.E., “Corrosion of carbon steels in high temperature acid chloride solutions”, Corrosion, No.10, (2012). 241

References

References

 Schmitt, G., and Horstemeier, M., Corrosion, Vol.6, No. 6112, NACE, (2006).  Schmitt, G., Bosch, C., Mueller, M. and Siegmund, G., Corrosion, No. 49, NACE, (2000).  Seo, T., Ryosuke, K. and Bunpei, S., “A convenient method for determining the concentration of hydrogen in water: use of methylene blue

with

colloidal

platinum”,

(http://www.medicalgasresearch

.com/content/2/1/1.html), (2012).  Sharma, K, S., “Corrosion Chemistry and Engineering”, 1st Edition, John-Wiley and Sons. (2012).  Sharqawy, M., Lienhard, J. and Zubair, S., “Thermophysical Properties of Sea Water: A review of Existing Correlations and Data”, No. 16, pp. (354-380), (2010).  Shaw, D.A. and Hanratty, T.J., Aiche, Vol. 23, No. 28, (1977).  Shrier, L.L. and Jarman, R.A. and Burstein, G.T., “Corrosion Metal/Environment Reactions”, 3rd Edition, Vol. 1, (2000).  Shrier, L.L. and Jarman, R.A. and Burstein, G.T., “Corrosion Control”, 3rd Edition, Vol. 2, (2000).  Silverman, D.C., “Conditions for Similarity of Mass Transfer Coefficients and Fluid Shear Stresses between the Rotating Cylinder Electrode and Pipe”, Corrosion, Vol. 6, No.6, p. 515, (2005).  Song, F.M., Kirk, D.W., Graydon, J.W. and Cormack, D.E., “Predicting Carbon Dioxide Corrosion of Bare Steel under an Aqueous Boundary Layer”, Corrosion, No. 60, pp.736-743, (2004).  Sridharan, V., M.Sc.Thesis, University of Saskatchewan, (2009).  Stern, M. and Geary, A.L., Electrochem. Soc., Vol.104, p.56 (1957).

242

References

References

 Sun, Y. and Nesic, S., “Kinetics of Corrosion Layer Formation, Part 1. Iron Carbonate Layers in Carbon Dioxide Corrosion”, Corrosion, No. 64, p.334, (2008).  Sun, Y., George, K. and Nesic, S., “The Effect of Cl- and Acetic Acid on Localized CO2 Corrosion in Wet Gas Flow”, Corrosion, No. 3327, NACE, (2003).  Tarantino, L., “Design and Analysis of Industrial Experiments”, Tetra Pak, (2010).  Tong, L., Yongjin, Y., Kewei, G. and Minxu, L., “Mechanism of protective film formation during CO2 corrosion of X65 pipeline steel”, Journal of University of Science and Technology, No. 15, pp. 702-719, (2008).  Tran, T., Brown. B. and Nesic, S., “Investigation of the Mechanism for Acetic Acid Corrosion of Mild Steel”, Corrosion, No. 2487, NACE, (2013).  Trethewey, K.R. & Chamberlain, J., “Corrosion for Science and Engineering”, 2nd Edition, Longman (1996).  Uhlig, H. H., “Corrosion Handbook”, 3rd Edition, John-Wiley and Sons, Inc., (2011).  Uhlig, H. H., “Handbook on Chemical Etching”, John-Wiley and Sons Inc., (2013).  Uhlig, H. H. and Winston, R. R., “Corrosion and Corrosion Control”, 4th Edition, John-Wiley and Sons Inc., (2008).  Villamizar, W., Casales, M., Gonzalez-Rodriguez, J.G. and Martinez, L., “CO2 Corrosion inhibition by hydroxyethyl, aminoethyl and amidoethyl imidazolines in Water-Oil Mixtures”, Solid State Electrochem., No.11, pp.619-629, (2007).  Wagner, C. and Traud, W., Z. Electrochem., Vol.44, p.391, (1938). 243

References

References

 Wang, H., Hongwei C.J.Y. and Jepson, W.P., Corrosion, No. 238, NACE, Houston, Texas (2002).  Wang, S. and Nesic, S., “ON Coupling CO2 Corrosion and Multiphase Flow Models”, Corrosion, No. 3631, NACE, Houston, Texas (2003).  Xiao, Y. and Nesic, S., “A Stochastic Prediction Model of Localized CO2 Corrosion”, Corrosion, No. 5057, NACE, (2005).  Yuli, P.A. and Mokhtar, C.I., “Application of Response Surface Design to Characterize CO2 Corrosion Mechanistically”, (Internet site: www.safan.com/ www.pm-pipeliner.safan.com), (2010).  Zhang, G., Chen, C., Lu, M., Chai, C. and Wu, Y. Mater. Chem. Phys. Vol. 105, No. 331, (2007).  Zhang, G., Lu, M., Chai, C. and Wu, Y., “Effect of HCO3concentration on CO2 corrosion in oil and gas fields”, Materials, Vol.13, No. 1, pp. 44-49, (2006).  Zhang, G. and Cheng, Y., “On the Fundamentals of Electrochemical Corrosion of X65 Steel in CO2-containing Formation Water in the Presence of Acetic Acid in Petroleum Production”, Corrosion Science, No.51,pp.(87-94),(2009).

244

Appendices

Appendices

Appendix (A)

Appendix (A) A.1 Sample of Calculation for Corrosion Rate (C.R) in Presence of Acetic Acid (Absence of the Protective Film Formation): For experimental run in presence of Acetic Acid at temperature 40 °C, pH 3, HAc acid concentration 1000 ppm and speed of rotation 1000 rpm (data from Table 4.2) the following values were determined: The corrosion rate for constant time 3h can be calculated from the following Equation. The surface area of cylindrical sample is 0.0013 m2. Therefore,

Where C.R: corrosion rate in (g/m2.day). W1: weight of a mild steel specimen before weight loss in (g). W2: weight of a mild steel specimen after weight loss in (g). A: surface area of the specimen in (m2). t: time in (day). C.R 

(25.4825  25.4744) g  48.5214 gmd day 2 0.0013 m x 3 h x 24 h

A.2 Sample of Calculation for Corrosion Rate (C.R) in Absence of Acetic Acid (Presence of the Protective Film Formation): For experimental run in absence of acetic acid at temperature 65 °C, pH 7.5 and speed of rotation 1000 rpm (data from Table 4.7) the following values were determined:

A1

Appendices

Appendix (A)

The corrosion rate for constant time 3h can be calculated from the following Equation. The surface area of cylindrical sample is 0.0013 m2. Therefore,

Where C.R: corrosion rate in (g/m2.day). W1: weight of a mild steel specimen before weight loss in (g). W2: weight of a mild steel specimen after weight loss in (g). A: surface area of the specimen in (m2). t: time in (day). C.R 

(25.2763  25.2737) g  15.8831 gmd day 2 0.0013 m x 3 h x 24 h

A2

Appendices

Appendix (B)

Appendix B B.1 Calculation of the Characteristic Equation Roots in Absence of Acetic Acid (Presence of the Protective Film Formation): Starting with the objective function (i.e. second order Equation describes the response surface in the region of the design under protective film formation.) An estimate of this Equation was: ̂

The first step is then to find S, from the regression Equation (the quadratic model) the optimized values are calculated by partial differentiating the above Equation with respect to x1, x2 & x3 and equating to zero, solving the three Equations from partial differentiation results in the optimal corrosion rate. The final optimal conditions are: Variable Temperature, °C pH Speed of Rotation, rpm

Code

Which is the point at which the response is stationary, which are the coordinates of S, substitution of these values in Equation (B.1) gives ̂

.

The second step is to determine the values of the coefficients θ11, θ22 and θ33. If B is a square matrix of order

and I is a unit matrix

of the same order, then the matrix: is called the characteristic matrix of A & θ is a parameter, such that the

B1

Appendices

| |

Appendix (B)

|

|

, θ is called the eign values of the matrix A.

[

]

[

[

]

]

then

[

]

[

]

Consider the determinant | | |

| =

|

| The upper left-hand corner and the lower right-hand corner terms

and the terms located between them are

,

respectively, where b11, b22 and b33 are the quadratic effects of the , and θ is an unknown quantity. These are called

fitted Equation

the diagonal terms of the determinant, and the remaining terms the nondiagonal terms; The latter are equal to

the interaction effects.

Multiplying out the below determinant and equating the result to zero, what is called the characteristic Equation is obtained. For present investigation this is given by:

The roots of this Equation are: . These are the values of the coefficients: ,

,

B2

Appendices

Appendix (B)

The other alternative θ

,θ

θ

could equally well have been taken. The result would only be to change the notation (the x1, x2, x3 - axes would have been called x3, x2, x1-axes, and vice versa).

B.2 Calculation of the Canonical Form in Absence of Acetic Acid (Presence of the Protective Film Formation): The canonical form of the fitted second-degree Equation

is

then: ̂ From the sign and relative magnitude of the coefficients , and from the position of S, the type of surface to be dealt with will now be clear.

B3

Appendices

Appendix (C)

Appendix (C) C.1 Examination of the Effective Variables (No Protective Film Formation): The effect of (F-test) is examined and a sample of calculation is given below: From the Equation (4.4) it is possible to compute estimated (predicted) ( ̂) and the experimental error (corresponding residuals) from STATISTICA software v.10, see Table 4.6: ̂ 1. The variance of coefficients is calculated from Equations (C.1 and C.2) Jeff Wu and Michael, [2009]: Sr    i / 2

… (C.1)

2

Sb  Sr /  x 2 2

2

... (C.2)

 = degree of freedom = N - N i coff

= number of experiments – number of coefficients = 20 – 15 = 5 Then from Equations (C.1) and (C.2) Sr 

21.8826  4.3765 5

Sb 

4.3765  0.2735 16

2

2

2. The coefficient of temperature (x1) from Equation (4.4) is equal to (b1 = 21.4734) then

= 461.1069

Let b2/Sb2 = Z = (461.1069/0.2735), then Z = 1685.9485 3. The F-value is calculated for 95% level of confidence with (1,5) degree of freedom from Tables, then: F-value = 6.61 Montgomery, [2005]

C1

Appendices

Appendix (C)

4. The significant of effects may be estimated by comparing the value of the ratio (Z) to the critical value of F0.95. Since Z > 6.61 then x1 is significant variable. 5. The results of the examination of the effective variables are given in Table 4.5.

C.2 Examination of the Effective Variables (In Presence of Protective Film Formation): The effect of (F-test) is examined and a sample of calculation is given below: From the Equation (4.11) it is possible to compute estimated (predicted) ( ̂) and the experimental error (corresponding residuals) from STATISTICA software v.10, see Table 4.11: ̂ 1. The variance of coefficients is calculated from Equations (C.1 and C.2) Jeff Wu and Michael, [2009]: Sr    i / 2

… (C.1)

2

Sb  Sr /  x 2 2

2

... (C.2)

 = degree of freedom = N - N i coff

= number of experiments – number of coefficients = 12 – 10 = 2 Then from Equations (C.1) and (C.2) Sr 

0.9082  0.4541 2

Sb 

0.4541  0.0568 8

2

2

2. The coefficient of temperature (x1) from Equation (4.11) is equal to (b1 = 8.9932) then

= 80.8776

Let b2/Sb2 = Z = (80.8776/0.0568), then Z = 1424.8590

C2

Appendices

Appendix (C)

3. The F-value is calculated for 95% level of confidence with (1,2) degree of freedom from Tables, then: F-value = 10.13 Montgomery, [2005] 4. The significant of effects may be estimated by comparing the value of the ratio (Z) to the critical value of F0.95. Since Z > 10.13 then x1 is significant variable. 5. The results of the examination of the effective variables are given in Table 4.10.

C3

Appendices

Appendix (D)

Appendix (D) D.1 Calculation of the Characteristic Equation Roots in Presence of Acetic Acid (Absence of the Protective Film Formation): Starting with the objective function (i.e. second order Equation describes the response surface in the region of the design under without the protective film formation.) An estimate of this Equation was: ̂

The first step is then to find S, from the regression Equation (the quadratic model) the optimized values are calculated by partial differentiating the above Equation (D.1) with respect to x1, x2, x3 & x4 and equating to zero, solving the four Equations from partial differentiation results in the optimal corrosion rate. The final optimal conditions are: Variable Temperature, °C pH HAc Acid Conc., ppm Speed of Rotation, rpm

Code

Which is the point at which the response is stationary which are the coordinates of S, substitution of these values in Equation (D.1) gives ̂

gmd.

The second step is to determine the values of the coefficients θ11, θ22, θ33 and θ44. If A is a square matrix of order matrix of the same order, then the matrix: D1

and I is a unit

Appendices

Appendix (D)

is called the characteristic matrix of A & θ is a parameter, such that the | |

|

|

[

, θ is called the eign values of the matrix A. ]

[

[

]

] then

[

[

]

]

Consider the determinant | | |

| =

|

|

The upper left-hand corner and the lower right-hand corner terms and the terms located between them are and

,

respectively, where b11, b22, b33 and b44 are the quadratic

effects of the fitted Equation

, and θ is an unknown quantity. These

are called the diagonal terms of the determinant, and the remaining terms the non-diagonal terms; the latter are equal to

the interaction

effects. Multiplying out the below determinant and equating the result to zero, what is called the characteristic Equation is obtained. For present investigation this is given by D2

Appendices

Appendix (D)

The roots of this Equation are

. These are the values of the coefficients indicates that the considered system is really a saddle system: , , ,

The other alternative , , ,

could equally well have been taken. The result would only be to change the notation (the x1, x2, x3 , x4 - axes would have been called x4, x3, x2, x1axes, and vice versa).

D.2 Calculation of the Canonical Form in Presence of Acetic Acid (Absence of the Protective Film Formation): The canonical form of the fitted second- degree Equation then: ̂

D3

is

Appendices

Appendix (D)

From the sign and relative magnitude of the coefficients , and from the position of S, the type of surface to be dealt with will now be clear.

D4

Appendices

Appendix (E)

Appendix (E) The Coefficients of β-Model: The coefficients of Equation (2.92) are evaluated using Non-Linear Estimation Method, and these values for no protective film formation and protective film formation are given below in Tables E.1 through E.4: Table E.1 The Coefficients of β-Model for No Protective Film Formation at Different Conditions. Temp. No pH (°C) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

40 50 60 40 50 60 40 50 60

HAc Conc. (ppm)


3

1000

1000

3

1000

1250

3

1000

1500

40

3

1000

50

3

1000

60

3

1000

1000 1250 1500 1000 1250 1500 1000 1250 1500

C1 µA/cm2.mV 9.1918 11.5414 21.8346 10.3978 12.5973 24.5845 11.2710 15.3334 25.0093 9.1918 10.3978 11.2710 11.5414 12.5973 15.3334 21.8346 24.5845 25.0093

E1

C2.102 C3.104 C4.105 2 2 2 3 µA/cm .mV µA/cm .mV µA/cm2.mV4 22.7707 27.3002 50.3236 27.3495 31.6388 60.0004 29.4036 38.3488 60.1874 22.7707 27.3495 29.4036 27.3002 31.6388 38.3488 50.3236 60.0004 60.1874

36.67 42.66 76.31 44.85 49.91 92.69 47.65 60.36 91.69 36.67 44.85 47.65 42.66 49.91 60.36 76.31 92.69 91.69

4.440 5.000 8.680 5.500 5.890 10.700 5.770 7.110 10.500 4.440 5.500 5.770 5.000 5.890 7.110 8.680 10.700 10.500

Appendices

Appendix (E)

Table E.2 The Coefficients of β-Model for No Protective Film Formation at Different Conditions. Temp. No pH (°C) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

40 50 60 40 50 60 40 50 60

5

5

5

40

5

50

5

60

5

Speed HAc of C1 C2.102 C3.104 C4.105 Conc. 2 2 2 2 3 Rotation µA/cm .mV µA/cm .mV µA/cm .mV µA/cm2.mV4 (ppm) (rpm) 7.4866 19.1461 30.39 3.610 3000 1000 7.9605 19.9330 31.54 3.740 9.9521 24.3860 37.96 4.420 7.9672 21.4167 35.89 4.500 3000 1250 8.5020 21.2054 33.48 3.960 10.0751 23.7584 35.20 3.900 8.3577 22.1478 36.37 4.460 3000 1500 9.1835 23.0290 36.32 4.290 10.1597 24.4533 36.95 4.170 1000 7.4866 19.1461 30.39 3.610 3000 1250 7.9672 21.4167 35.89 4.500 1500 8.3577 22.1478 36.37 4.460 1000 7.9605 19.9330 31.54 3.740 3000 1250 8.5020 21.2054 33.48 3.960 1500 9.1835 23.0290 36.32 4.290 1000 9.9521 24.3860 37.96 4.420 3000 1250 10.0751 23.7584 35.20 3.900 1500 10.1597 24.4533 36.95 4.170

E2

Appendices

Appendix (E)

Table E.3 The Coefficients of β-Model for Protective Film Formation at Different Conditions. Speed Temp. of C1 C2.103 C3.105 C4.106 No pH 2 2 2 2 3 (°C) Rotation µA/cm .mV µA/cm .mV µA/cm .mV µA/cm2.mV4 (rpm) 1 65 2.6288 63.4180 94.800 10.600 2 70 7.5 1000 3.7798 87.1970 127.800 14.000 3 75 5.9602 134.1380 194.300 21.100 4 65 2.8944 70.1480 105.300 11.800 5 70 7.5 1250 4.3204 99.3340 145.600 16.000 6 75 6.0684 135.1680 195.000 2.1100 7 65 3.4670 82.2070 122.600 13.700 8 70 7.5 1500 4.8834 111.8590 164.200 18.100 9 75 6.2082 139.1440 200.800 21.700 10 1000 2.6288 63.4180 94.800 10.600 11 65 7.5 1250 2.8944 70.1480 105.300 11.800 12 1500 3.4670 82.2070 122.600 13.700 13 1000 3.7798 87.1970 127.800 14.000 14 70 7.5 1250 4.3204 99.3340 145.600 16.000 15 1500 4.8834 111.8590 164.200 18.100 16 1000 5.9602 134.1380 194.300 21.100 17 75 7.5 1250 6.0684 135.1680 195.000 21.100 18 1500 6.2082 139.1440 200.800 21.700

E3

Appendices

Appendix (E)

Table E.4 The Coefficients of β-Model for Protective Film Formation at Different Conditions. Speed Temp. of C1 C2.103 C3.105 C4.105 No pH 2 2 2 2 3 (°C) Rotation µA/cm .mV µA/cm .mV µA/cm .mV µA/cm2.mV4 (rpm) 1 65 2.0538 50.011 75.100 0.843 2 70 8.5 1000 3.7251 86.483 127.500 1.410 3 75 5.2862 118.863 171.300 1.850 4 65 2.4082 58.833 89.000 1.010 5 70 8.5 1250 3.9639 90.622 133.000 1.460 6 75 5.4502 122.053 176.800 1.920 7 65 2.7259 65.598 98.200 1.100 8 70 8.5 1500 4.0839 94.454 138.300 1.520 9 75 5.4558 122.822 177.300 1.920 10 1000 2.0538 50.011 75.100 0.843 11 65 8.5 1250 2.4082 58.833 89.000 1.010 12 1500 2.7259 65.598 98.200 1.100 13 1000 3.7251 86.483 127.500 1.410 14 70 8.5 1250 3.9639 90.622 133.000 1.460 15 1500 4.0839 94.454 138.300 1.520 16 1000 5.2862 118.863 171.300 1.850 17 75 8.5 1250 5.4502 122.053 176.800 1.920 18 1500 5.4558 122.822 177.300 1.920

E4

Appendices

Appendix (F)

Appendix (F) The Derivation of β-Model: Equation (2.71) which given in chapter two can be written in the form of the degree of polynomial as; i   Co  C1  C2 2  C3 3  ...  Cn n

… (2.92)

Where Co, C1, C2, …Cn are constants. Equation (2.92) is a non-linear Equation of   i  relationship. Equation (2.92) can be written in the form of the Maclaurin Formula as; 1   2i  i  i        2 2!      0

 1   ni   2  ...   n n!     0

   n   0

… (2.93)

The expressions of currents as a function of potential are given by Equation (2.71) as;    2.303   exp     bc    2.303      i  icorr exp     b a  1     exp   2.303       b c      

… (2.71)

So,    2.303   2.303    exp   exp     bc  bc  i  2.303  2.303   2.303          icorr  exp       2.303    2.303     b a   b c    ba 1     exp         1     exp   b     b c c         2.303 2.303 2.303   i     icorr     bc bc     0  ba

1  1  C1  2.303icorr   1    ba bc 

F1

… (2.95)

Appendices

Appendix (F)

   2.3   exp   2 2 2    2.3   bc   2.3  exp  2.3    2.3      3  b  b    b a   b   2.3   a   c    c        1     exp      bc     2i   2   icorr   2 3  2.3   2.3         exp   exp   2    2.3   bc   bc  2       2  2 3 b     c    2.3    2.3   1     exp        1     exp     b    b    c  c    

2 2  2.3  2  2.3  2  2.3   2.3  2    2i       3    2     2   icorr  b b b b     0  a   c    c   c 

 1  2C 2  5.303i corr   b a

2

2

  1      1  3  2 2   bc 





 

… (2.96)

   2.3     exp   3 3  2.3  3   2.3   2.3   2.3   bc   exp           7 bc   2.3    b a    b a   b c        1     exp   b   c      2 3    2.3   2.3         exp exp  3   3    2.3   i    bc   bc  2    3   icorr    12   2 3 b      c    2.3    2.3     1     exp        1     exp     b    c   bc      4    2.3     exp     2.3  3  b c   3    6  4   bc     2.3   1     exp             bc  

F2

Appendices

Appendix (F)

3 3 3 3 3   2.3   2.3  2  2.3  3    3i   2.3   2.3        7    12    6     3   icorr  b b    0    bc   bc   bc   a   c  

 1  6C3  12.214icorr   b a

3

  1      bc

3

  1  7   12 2  6 3 







… (2.97)



   2.3η     exp   4 4  2.3  4  b  2.3η   2.3    2.3 c         exp    15   b b b    b a     2.3η   a   c   c    1     exp      bc     2 3    2.3η   2.3η      exp   exp   4 4   b b  2.3   i  c  c    2    4   icorr    50   2 3 b      c      2.3η    2.3η   1     exp        1     exp         bc   bc     4 5    2.3η   2.3η      exp   exp   4 4   b b  2.3    2.3 c  c    3 4    60    24    4 5 b b    c    c      1     exp   2.3η   1     exp   2.3η      b   b      c  c     

4 4 4 4  2.3  4  2.3  4  2.3   2.3  2  2.3  3  2.3  4    4i   4   icorr       15    50    60    24       0  b a   b c    bc   bc   bc   bc 

  1  24C 4  28.13icorr  b   a

4

4

  1      1  15  50 2  60 3  24 4   bc 



F3



 … (2.98)  

Appendices

Appendix (G)

Appendix (G) Substitution Procedure: The following Equations were obtained by comparison of Equation (2.92) with Equation (2.93) which given in chapter two. The polynomial coefficients can be written as;  1   1   C1  2.303icorr     1     b a   b c   2 2    1   1   2C 2  5.303icorr      1  3  2 2  b b    a   c   3 3    1   1   6C3  12.214icorr      1  7   12 2  6 3  b b    a   c   4 4   1   1  24C 4  28.13icorr      1  15  50 2  60 3  24 4 b b   a   c 



… (2.95)





… (2.96)





… (2.97) 

  

… (2.98)

For each experiment the values of C1-C4 were obtained and can be substituted in Equations (2.95 through 2.98). By using MATLAB v.12 (Matlab Professional) Software Computer Program, the sets of above Equation can be solved, the procedure was as follows: 1. From Equation (2.95), the value of ba can be obtained in term of (bc, icorr ,  ), i.e., ba  f bc , icorr ,  

… (G.1)

2. Substitute Equation (G.1) in Equation (2.96), and then the value of bc can be obtained in term of ( icorr ,  ), so; bc  f icorr ,  

… (G.2)

3. Substitute Equations (G.1) and (G.2) in Equations (2.97) and (2.98). Two complex Equations of different orders can be obtained in term of icorr and  , so; f1 icorr ,    0

G1

… (G.3)

Appendices

Appendix (G) f 2 icorr ,    0

… (G.4)

4. Assume icorr , (The assumed values were around the values of icorr obtained by McLaughlin method), Equations (G.3) and (G.4) can be re-write as: f 1    0

… (G.5)

f 2    0

… (G.6)

5. From Equation (G.5), and by trial and error, the values of  can be obtained (Note that, 0    1 ). 6. Substitute the value of  obtained in Equation (G.6), if f 2    0 , then the value of icorr assumed was right, if not, then assume another value of icorr and repeat the above procedure. 7. After computing the values of  and icorr , the values of bc and ba can be obtained from Equation (G.2) and (G.1) respectively.

G2

Appendices

Appendix (H)

Appendix (H) Mass Transfer Correction Factors: Equation (2.91) which given in chapter two can used to evaluate the values of mass transfer correction factors, these values are given below for presence and absence of acetic acid (absence and presence of the protective film formation) at different conditions; Table H.1 Mass Transfer Correction Factors in Presence of Acetic Acid (Absence of Protective Film Formation) at pH 3, 1000 ppm HAc, 1000 rpm and Different Temperatures. η (mV) 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

λ, T = 40 °C 0.9539 0.9095 0.8668 0.8257 0.7862 0.7484 0.7119 0.6772 0.6438 0.6118 0.5813 0.5521 0.5242 0.4976 0.4721

λ, T = 50 °C 0.9653 0.9317 0.8992 0.8677 0.8372 0.8077 0.7792 0.7515 0.7248 0.6989 0.6739 0.6498 0.6265 0.6039 0.5821

H1

λ, T = 60 °C 0.9725 0.9456 0.9193 0.8936 0.8685 0.8439 0.8200 0.7967 0.7739 0.7516 0.7299 0.7088 0.6882 0.6681 0.6486

Appendices

Appendix (H)

Table H.2 Mass Transfer Correction Factors in Presence of Acetic Acid (Absence of Protective Film Formation) at pH 3,1000 ppm HAc, 1250 rpm and Different Temperatures. η (mV) 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

λ, T = 40 °C 0.9764 0.9526 0.9287 0.9046 0.8805 0.8563 0.8320 0.8079 0.7838 0.7598 0.7359 0.7122 0.6888 0.6656 0.6427

λ, T = 50 °C 0.9896 0.9790 0.9683 0.9573 0.9461 0.9348 0.9233 0.9116 0.8997 0.8877 0.8755 0.8632 0.8507 0.8382 0.8254

λ, T = 60 °C 0.9918 0.9834 0.9749 0.9663 0.9575 0.9486 0.9396 0.9304 0.9211 0.9117 0.9022 0.8925 0.8828 0.8729 0.8629


λ, T = 40 °C 0.9798 0.9593 0.9385 0.9175 0.8963 0.8749 0.8534 0.8318 0.8101 0.7883 0.7666 0.7449 0.7232 0.7017 0.6802

λ, T = 50 °C 0.9900 0.9798 0.9695 0.9589 0.9482 0.9373 0.9263 0.9151 0.9037 0.8922 0.8805 0.8687 0.8568 0.8447 0.8325

H2

λ, T = 60 °C 0.9919 0.9836 0.9752 0.9667 0.9580 0.9493 0.9403 0.9313 0.9222 0.9129 0.9035 0.8939 0.8844 0.8746 0.8648

Appendices

Appendix (H)


λ, T = 40 °C 0.9800 0.9597 0.9392 0.9185 0.8975 0.8764 0.8551 0.8338 0.8123 0.7908 0.7693 0.7479 0.7264 0.7051 0.6838

λ, T = 50 °C 0.9864 0.9727 0.9588 0.9448 0.9306 0.9163 0.9018 0.8873 0.8727 0.8579 0.8431 0.8283 0.8133 0.7984 0.7834

λ, T = 60 °C 0.9902 0.9803 0.9703 0.9601 0.9499 0.9395 0.9289 0.9184 0.9076 0.8968 0.8859 0.8749 0.8638 0.8526 0.8414


λ, T = 40 °C 0.9769 0.9535 0.9299 0.9062 0.8824 0.8585 0.8346 0.8106 0.7868 0.7629 0.7393 0.7157 0.6924 0.6693 0.6464

λ, T = 50 °C 0.9866 0.9731 0.9594 0.9456 0.9316 0.9175 0.9033 0.8891 0.8747 0.8602 0.8457 0.8311 0.8165 0.8018 0.7871

H3

λ, T = 60 °C 0.9955 0.9908 0.9861 0.9812 0.9762 0.9711 0.9659 0.9605 0.9551 0.9495 0.9438 0.9380 0.9321 0.9261 0.9199

Appendices

Appendix (H)


λ, T = 40 °C 0.9802 0.9602 0.9398 0.9191 0.8982 0.8771 0.8558 0.8343 0.8128 0.7911 0.7694 0.7477 0.7260 0.7044 0.6829

λ, T = 50 °C 0.9899 0.9796 0.9691 0.9585 0.9476 0.9366 0.9254 0.9141 0.9026 0.8909 0.8791 0.8671 0.8551 0.8428 0.8305

λ, T = 60 °C 0.9964 0.9927 0.9889 0.9850 0.9811 0.9769 0.9728 0.9685 0.9641 0.9596 0.9550 0.9503 0.9455 0.9405 0.9355

Table H.7 Mass Transfer Correction Factors in Absence of Acetic Acid (Presence of Protective Film Formation) at pH 7.5, 1000 rpm and Different Temperatures. η (mV) 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

λ, T = 65 °C 0.9803 0.9604 0.9402 0.9199 0.8993 0.8786 0.8578 0.8369 0.8159 0.7949 0.7739 0.7528 0.7319 0.7109 0.6902

λ, T = 70 °C 0.9838 0.9676 0.9513 0.9349 0.9185 0.9021 0.8857 0.8693 0.8528 0.8364 0.8201 0.8037 0.7875 0.7713 0.7551

H4

λ, T = 75 °C 0.9823 0.9647 0.9472 0.9297 0.9123 0.8950 0.8778 0.8608 0.8438 0.8269 0.8102 0.7937 0.7772 0.7609 0.7449

Appendices

Appendix (H)


λ, T = 65 °C 0.9802 0.9601 0.9397 0.9192 0.8985 0.8776 0.8566 0.8354 0.8142 0.7930 0.7718 0.7506 0.7294 0.7083 0.6874

λ, T = 70 °C 0.9827 0.9654 0.9481 0.9308 0.9135 0.8962 0.8789 0.8617 0.8445 0.8274 0.8104 0.7934 0.7766 0.7599 0.7432

λ, T = 75 °C 0.9812 0.9626 0.9441 0.9257 0.9075 0.8895 0.8716 0.8539 0.8364 0.8191 0.8019 0.7850 0.7683 0.7518 0.7355


λ, T = 65 °C 0.9728 0.9457 0.9187 0.8918 0.8652 0.8388 0.8126 0.7868 0.7612 0.7361 0.7113 0.6869 0.6629 0.6394 0.6163

λ, T = 70 °C 0.9792 0.9584 0.9378 0.9173 0.8969 0.8767 0.8567 0.8368 0.8171 0.7976 0.7783 0.7592 0.7404 0.7218 0.7034

H5

λ, T = 75 °C 0.9830 0.9661 0.9493 0.9325 0.9157 0.8991 0.8825 0.8660 0.8496 0.8334 0.8172 0.8012 0.7853 0.7695 0.7539

Appendices

Appendix (H)

Table H.10 Mass Transfer Correction Factors in Absence of Acetic Acid (Presence of Protective Film Formation) at pH 8.5,1000 rpm and Different Temperatures. η (mV) 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

λ, T = 65 °C 0.9817 0.9631 0.9442 0.9249 0.9055 0.8859 0.8659 0.8459 0.8258 0.8055 0.7851 0.7647 0.7442 0.7237 0.7033

λ, T = 70 °C 0.9844 0.9687 0.9529 0.9370 0.9211 0.9052 0.8892 0.8732 0.8572 0.8412 0.8252 0.8093 0.7934 0.7775 0.7617

λ, T = 75 °C 0.9855 0.9709 0.9564 0.9418 0.9272 0.9126 0.8980 0.8834 0.8689 0.8544 0.8399 0.8254 0.8110 0.7967 0.7824


λ, T = 65 °C 0.9801 0.9599 0.9396 0.9189 0.8982 0.8772 0.8561 0.8349 0.8136 0.7923 0.7709 0.7497 0.7284 0.7072 0.6862

λ, T = 70 °C 0.9789 0.9580 0.9372 0.9165 0.8960 0.8757 0.8555 0.8356 0.8158 0.7963 0.7769 0.7579 0.7390 0.7204 0.7021

H6

λ, T = 75 °C 0.9828 0.9657 0.9486 0.9317 0.9148 0.8981 0.8815 0.8650 0.8487 0.8324 0.8163 0.8004 0.7846 0.7689 0.7535

Appendices

Appendix (H)


λ, T = 65 °C 0.9775 0.9547 0.9319 0.9089 0.8858 0.8627 0.8396 0.8165 0.7935 0.7706 0.7477 0.7251 0.7026 0.6803 0.6583

λ, T = 70 °C 0.9865 0.9728 0.9591 0.9452 0.9313 0.9172 0.9031 0.8889 0.8746 0.8602 0.8459 0.8314 0.8169 0.8025 0.7880

H7

λ, T = 75 °C 0.9869 0.9737 0.9605 0.9473 0.9339 0.9207 0.9074 0.8941 0.8808 0.8675 0.8542 0.8409 0.8276 0.8144 0.8012

Appendices

Appendix (I)

Appendix (I) Calculation of Activation Enthalpy (ΔH*) and Activation Entropy (ΔS*) From Transition State Equation: pH pH pH pH pH

Log (Corr. Rate/T), (gmd)K -1

-0.4

= = = = =

3 ; log(C.R) / T) = 5.4243 - 1.9545 x (1000/T); R = 0.99 log(C.R/T)atpH3 = 5.4243-1.9545*x log(C.R/T)atpH3.5 = 5.1514-1.8989*x 3.5;log(C.R) / T) = 5.1514 - 1.8989 x (1000/T);R = 1.00 log(C.R/T)atpH4 = 4.2758-1.6444*x 4 ; log(C.R) / T) = 4.2758 - 1.6444 x (1000/T); R = 1.00 log(C.R/T)atpH4.5 = 2.3919-1.0567*x 4.5 ;log(C.R) / T)= 2.3919 - 1.0567 x (1000/T); R = 0.99 log(C.R/T)atpH5 = 1.9551-0.9262*x 5 ; log(C.R) / T) = 1.9551 - 0.9262 x (1000/T); R = 0.99 pH = pH = pH = pH = pH =

-0.5 -0.6

3 3.5 4 4.5 5

-0.7 -0.8 -0.9 -1.0 -1.1 2.98

3.00

3.02

3.04

3.06

3.08

3.10

3.12

3.14

3.16

3.18

3.20

3.22

(1000/T) (K -1 )

Fig. I.1 Transition State Equation Plot for API X65 Mild Steel in CO2 Saturated, 3.5 wt% NaCl Solutions in Presence of Acetic Acid and without Protective Film Formation at 1000 ppm HAc & 1000 rpm.

Log (Corr. Rate/T), (gmd)K-1

-0.3

pH = pH = pH = pH = pH =

3 ; log(C.R)log(C.R/T)atpH3 / T) = 5.3802= -5.3802-1.93*x 1.9300 x (1000/T); R = 1.00 log(C.R/T)atpH3.5 = 5.0996-1.8694*x 3.5;log(C.R) / T) = 5.0996 - 1.8694 x (1000/T);R = 1.00 log(C.R/T)atpH4 = 4.2458-1.6197*x 4 ; log(C.R) / T) = 4.2458 - 1.6197 x (1000/T); R = 0.99 log(C.R/T)atpH4.5 = 2.8728-1.1948*x 4.5 ;log(C.R) / T)= 2.8728 - 1.1948 x (1000/T); R = 1.00 log(C.R/T)atpH5 = 2.4767-1.0787*x 5 ; log(C.R) / T) = 2.4767 - 1.0787 x (1000/T); R = 0.99 pH = 3 pH = 3.5 pH = 4 pH = 4.5 pH = 5

-0.4 -0.5 -0.6 -0.7 -0.8 -0.9 -1.0 2.98

3.00

3.02

3.04

3.06

3.08

3.10

3.12

3.14

3.16

3.18

3.20

3.22

(1000/T) (K -1 )


I1

Appendices

Appendix (I) pH = 3 ; log(C.R) / T ) = 5.0665 - 1.8330 x (1000/T ); R = 1.00 log(C.R/T)atpH3 = 5.0665-1.833*x pH = 3.5;log(C.R) / T ) = 4.7106 - 1.7493 x (1000/T );R = 0.99 log(C.R/T)atpH3.5 = 4.7106-1.7493*x log(C.R/T)atpH4 = 3.8069-1.4848*x pH = 4 ; log(C.R) / T ) = 3.8069 - 1.4848 x (1000/T ); R = 1.00 = 2.1827-0.9811*x pH = 4.5 log(C.R/T)atpH4.5 ;log(C.R) / T )= 2.1827 - 0.9811 x (1000/T ); R = 1.00 log(C.R/T)atpH5 = 1.6865-0.8321*x pH = 5 ; log(C.R) / T ) = 1.6865 - 0.8321 x (1000/T ); R = 1.00


-0.4

pH = 3 pH = 3.5 pH = 4 pH = 4.5 pH = 5

-0.5 -0.6 -0.7 -0.8 -0.9 -1.0 2.98

3.00

3.02

3.04

3.06

3.08

3.10

3.12

3.14

3.16

3.18

3.20

3.22

(1000/T) (K -1 )


pH = pH = pH = pH = pH =

Log (Corr. Rate/T), (gmd)K

-1

-0.3

log(C.R/T)atpH3 = 5.0448-1.816*x 3 ; log(C.R) / T) = 5.0448 - 1.8160 x (1000/T); R = 1.00 log(C.R/T)atpH3.5 = 4.7176-1.7391*x 3.5;log((C.R) / T) = 4.7176 - 1.7391 x (1000/T);R = 0.99 log(C.R/T)atpH4 = 3.9008-1.5008*x 4 ; log(C.R)/ T) = 3.9008 - 1.5008 x (1000/T); R = 1.00 log(C.R/T)atpH4.5 = 2.3688-1.0236*x 4.5 ;log(C.R) / T)= 2.3688 - 1.0236 x (1000/T); R = 0.99 log(C.R/T)atpH5 = 1.8074-0.8513*x 5 ; log(C.R) / T) = 1.8074 - 0.8513 x (1000/T); R = 0.99

pH = 3 pH = 3.5 pH = 4 pH = 4.5 pH = 5

-0.4 -0.5 -0.6 -0.7 -0.8 -0.9 -1.0 2.98

3.00

3.02

3.04

3.06

3.08

3.10

3.12

3.14

3.16

3.18

3.20

3.22

(1000/T) (K -1 )

Fig. I.4 Transition State Equation Plot for API X65 Mild Steel in CO2 Saturated,3.5 wt% NaCl Solutions in Presence of Acetic Acid and without Protective Film Formation at 3000 ppm HAc & 1500 rpm.

I2

Appendices

Appendix (I)


-1.0

pH = 7.5 ;log(C.R/T ) = 12.3613 - 4.7056 x (1000/T );R = 1.00 log(C.R/T)atpH7.5 = 12.3613-4.7056*x log(C.R/T)atpH7.75 = 8.7809-3.4232*x pH = 7.75;log(C.R/T ) = 8.7809 - 3.4232 x (1000/T ); R = 0.99 log(C.R/T)atpH8 = 8.0191-3.1512*x pH = 8 ; log(C.R/T ) = 8.0191 - 3.1512 x (1000/T ); R = 0.99 log(C.R/T)atpH8.25 = 9.7315-3.7622*x pH = 8.25;log(C.R/T ) = 9.7315 - 3.7622 x (1000/T ); R = 1.00 log(C.R/T)atpH8.5 = 15.6885-5.8954*x pH = 8.5 ;log(C.R/T ) = 15.6885 - 5.8954 x (1000/T );R = 1.00

-1.1 -1.2 -1.3 -1.4 -1.5 -1.6 -1.7 -1.8 2.87

pH = 7.5 pH = 7.75 pH = 8 pH = 8.25 pH = 8.5 2.88

2.89

2.90

2.91

2.92

2.93

2.94

2.95

2.96

2.97

(1000/T) (K -1 )

Fig. I.5 Transition State Equation Plot for API X65 Mild Steel in CO2 Saturated, 3.5 wt% NaCl Solutions in Absence of Acetic Acid and with Protective Film Formation at 1000 rpm.


-0.9

pH = 7.5 log(C.R/T)atpH7.5 ;log(C.R/T ) = 8.2953 - 3.2781 x (1000/T );R = 1.00 = 8.2953-3.2781*x log(C.R/T)atpH7.75 = 6.3675-2.5735*x pH = 7.75;log(C.R/T ) = 6.3675 - 2.5735 x (1000/T );R = 1.00 log(C.R/T)atpH8 = 6.1025-2.475*x pH = 8 ; log(C.R/T ) = 6.1025 - 2.4750 x (1000/T ); R = 0.99 log(C.R/T)atpH8.25 = 6.9579-2.7903*x pH = 8.25;log(C.R/T ) = 6.9579 - 2.7903 x (1000/T );R = 1.00 log(C.R/T)atpH8.5 = 10.4905-4.0738*x pH = 8.5 ;log(C.R/T ) = 10.4905 - 4.0738 x (1000/T );R = 1.00

-1.0 -1.1 -1.2 -1.3 -1.4 -1.5 -1.6 2.87

pH = 7.5 pH = 7.75 pH = 8 pH = 8.25 pH = 8.5 2.88

2.89

2.90

2.91

2.92

2.93

2.94

2.95

2.96

2.97

(1000/T) (K -1 )

Fig. I.6 Transition State Equation Plot for API X65 Mild Steel in CO2 Saturated, 3.5 wt% NaCl Solutions in Absence of Acetic Acid and with Protective Film Formation at 1500 rpm.

I3

Appendices

Appendix (J)

Appendix (J) J.1 API X65 Mild Steel Protective Film Thickness: Table J.1:- Experimental Runs for API X65Mild Steel in CO2-Saturated, 3.5 wt% NaCl Solution in Absence of Acetic Acid (Presence of the Protective Film Formation) by using the Optimum Conditions. Run No.

1*

W1 (g)

Wf (g)

Wwf (g)

WFilm WM 2 (mg/dm ) (mg/dm2)

RF

Film Thickness (Ft) µm

24.6301 24.6312 24.6265 24.8305 24.8311 24.8263

36.15 36.92

27.69 32.31

1.31 1.14

91.297 93.240

1Av. 24.6359 24.6367 24.6320

36.15 36.41

30.00 30.00

1.21 1.21

91.297 91.944

1** 1*** 24.4471 24.4479 24.4432

Where : WM: Weight of mild steel reacted in (mg/dm2) = (W1 - Wwf )/A RF: Film ratio = WFilm/ WM

J.2 API X65 Mild Steel Protective Film Porosity Calculations: Film porosity (ε) is calculated using the following expression:

where thickness of porous film is found from optical microscopy image of the cross-section as described in Figure 4.116, and thickness of 100% film is found out using the following calculations:

where A is the surface area in dm2 of the specimen and vol. of 100% film is found using:

J1

Appendices

where

Appendix (J)

is the weight of the specimen with the film in gm,

is the

weight of the specimen after film removal (after cleaning) using chemical etching solution in gm,

is the density of a 100% FeCO3 film (3.96

gm/cm3). The surface area of the specimens used for weight loss measurements was 0.13 dm2.

J.3 X-Ray Diffraction: Table J.2:- X-Ray Diffraction Pattern of Corroded API X65 Mild Steel Surface Specimen in CO2-Saturated, 3.5 wt% NaCl Solution at the Optimum Conditions in (a) Presence of Acetic Acid (Absence of Protective Film Formation). Component

2θ (deg)

d-spacing (Å)

Intensity

Fe3C FeCO3

44.786 65.038

2.065 2.144

100 20

Table J.3:- X-Ray Diffraction Pattern of Corroded API X65 Mild Steel Surface Specimen in CO2-Saturated, 3.5 wt% NaCl Solution at the Optimum Conditions in (b) Absence of Acetic Acid (Presence of Protective Film Formation). Component

2θ (deg)

d-spacing (Å)

Intensity

FeCO3

24.023, 32.7765, 38.265, 43.399, 45.002, 46.1943, 52.243

1.738, 2.795, 1.965, 1.965, 1.965, 1.506, 1.732

30, 90, 20,20,20,14, 35

J2

Appendices

Appendix (C)

Appendix (K) Experiments in a Simulated Brine Compared to 3.5 wt% NaCl Solutions Saturated with CO2 Gas in Presence and Absence of Acetic Acid at Different Experimental Conditions: 60

WL


50

40

30

20

10

0

3.5 wt% NaCl

Simulated Brine

Fig. K.1 The Effect of 3.5 wt% NaCl and Simulated Brines Solutions on the Corrosion Rate of API X65 Mild Steel in CO2 Saturated Solutions (40 ºC, pH 3, 1000 ppm HAc & 1000 rpm) by Weight Loss Technique.

50

WL 45


40 35 30 25 20 15 10 5 0

3.5 wt% NaCl

Simulated Brine


K1

Appendices

Appendix (C)

140

WL


120

100

80

60

40

20

0

3.5 wt% NaCl

Simulated Brine

Fig.K.3 The Effect of 3.5 wt% NaCl and Simulated Brines Solutions on the Corrosion Rate of API X65 Mild Steel in CO2 Saturated Solutions (60 ºC, pH 3, 1000 ppm HAc & 1500 rpm) by Weight Loss Technique.

70

WL


60

50

40

30

20

10

0

3.5 wt% NaC l

Simulate d Brine


K2

Appendices

Appendix (C)

70

WL


60

50

40

30

20

10

0

3.5 wt% NaC l

Simulate d Brine


20 18

WL


16 14 12 10 8 6 4 2 0

3.5 wt% NaCl

Simulated Brine

Fig. K.6 The Effect of 3.5 wt% NaCl and Simulated Brines Solutions on the Corrosion Rate of API X65Mild Steel in CO2 Saturated Solutions (65 ºC, pH 7.5, 0 ppm HAc & 1000 rpm) by Weight Loss Technique.

K3

Appendices

Appendix (C)

16

14

WL


12

10

8

6

4

2

0

3.5 wt% NaCl

Simulated Brine

Fig. K.7 The Effect of 3.5 wt% NaCl and Simulated Brines Solutions on the Corrosion Rate of API X65 Mild Steel in CO2 Saturated Solutions (65 ºC, pH 8.5, 0 ppm HAc & 1000 rpm) by Weight Loss Technique.

45

40

WL


35

30

25

20

15

10

5

0

3.5 wt% NaCl

Simulated Brine

Fig.K.8 The Effect of 3.5 wt% NaCl and Simulated Brines Solutions on the Corrosion Rate of API X65 Mild Steel in CO2 Saturated Solutions (75 ºC, pH 7.5, 0 ppm HAc & 1500 rpm) by Weight Loss Technique.

K4

Appendices

Appendix (C)

40

35

WL


30

25

20

15

10

5

0

3.5 wt% NaC l

Simulate d Brine

Fig. K.9 The Effect of 3.5 wt% NaCl and Simulated Brines Solutions on the Corrosion Rate of API X65 Mild Steel in CO2 Saturated Solutions (75 ºC, pH 8.5, 0 ppm HAc & 1500 rpm) by Weight Loss Technique.

24 22 20

WL


18 16 14 12 10 8 6 4 2 0

3.5 wt% NaCl

Simulated Brine


K5

‫الخالصة‬


‫تم أيضا إيجاد قيم مقاومة األستقطاب )‪ (polarization resistance, Rp‬ووجد أن قيمة‬ ‫‪ Rp‬تزداد بتناقص درجة الحرارة وسرعة التدوير بغياب وبوجود تكوين الطبقة الحامية‪ .‬ووجد‬ ‫أن قيم )‪ (Rp‬أكبر بغياب حامض الخليك من وجوده نتيجة لتكوين الطبقة الحامية‪.‬‬ ‫وتم إيجاد قيم معامل التصحيح ألنتقال المادة (‪ )λ‬بغياب وبوجود تكوين الطبقة الحامية‪.‬‬ ‫ووجد أن قيمة (‪ )λ‬تقترب قيمتها من واحد عند قيم واطئة لفوق الجهد ووجد قيم (‪ )λ‬تتناقص‬ ‫بزيادة قيم فوق الجهد‪ .‬بصورة عامة بغياب حامض الخليك وجد أن قيم (‪ )λ‬تقترب فيما بينها‬ ‫وتتوحد قيمها مقارنة بوجود حامض الخليك عند أختالف درجات الحرارة وسرع التدوير نتيجة‬ ‫لحاجز األنتشار المتكون بسبب تكوين الطبقة الحامية يعجل بأجراءات تقييد أنتقال نواتج التفاعل‬ ‫من السطح‪.‬‬ ‫كما وجد بأن تيار الهيدروجين المحدد أرتبط في ظروف الجريان المضطرب بوجود‬ ‫حامض الخليك وبغيابه ‪.‬‬

‫‪ .3‬تقنيااات تمييااا خصااائح السااط المتآكاال ( ‪Characterization of the‬‬ ‫)‪Corroded Surface Techniques‬‬ ‫أجريت دراسات مقارنة بين النماذج مين الحدييد المطياوع (‪ )API X65‬المتآكلية فيي تقنيية‬ ‫الفقدان بالوزن بوجود وبغياب تأثيرحامض الخليك عند الظروف المثالية ونماذج من الميواد أعياله‬ ‫بيييدون تآكيييل (‪° 43¸4‬م ‪ ,‬أس هييييدروجيني ‪2118¸3 , 4¸8‬جيييزء بيييالمليون حيييامض الخلييييك و‬ ‫‪ 1296¸6‬دورة بالدقيقة) وبغياب حامض الخلييك (‪° 68¸1‬م ‪ ,‬أس هييدروجيني ‪ 1¸9‬و ‪1423¸8‬‬ ‫دورة بالدقيقة) وتضمنت دراسات المقارنة أجراء تحاليل لسمك ومسيامية الطبقية الحاميية ‪ ,‬دراسية‬ ‫خشييييونة وصييييالدة السييييطح )‪ , (VMH‬فحييييص البنييييية المجهرييييية بأسييييتخدام المجهرالفيييياحص‬ ‫األلكترونيي )‪ (SEM‬وبأسيتخدام المجهرالضيوئي )‪ (CMOMT‬وحييود أشيعة أكيس )‪.(XRD‬‬ ‫فييي العمييوم لييوح بوجييود حييامض الخليييك تكونييت طبقيية مسييامية (كاربيييد الحديييد وكاربونييات‬ ‫الحديدوز) لكن بغياب حامض الخليك تكونت طبقة كثيفة تماما (كاربونات الحديدوز فقط) وليوح‬ ‫أيضا بأن قيم الخشونة والصالدة للعينات الناتجة من التآكل بغيياب حيامض الخلييك أكبير بالمقارنية‬ ‫مع العينات الناتجة من التآكل بوجود حامض الخليك‪.‬‬

‫‪-3-‬‬



‫صممت تجارب التآكل لتشكيل تعبير رياضي من المرتبة الثانية بأستخدام تصميم‬ ‫فكتوريل الكلي العملي )‪:(FFED‬‬ ‫أ‪ .‬تجارب بأربعة متغيرات ( تأثير درجة الحرارة ‪ ,‬حامضية المحلول ‪ ,‬تركيز حامض‬ ‫الخليك وسرعة التدوير)‪.‬‬ ‫ب‪ .‬تجارب بثالثة متغيرات ( تأثير درجة الحرارة ‪ ,‬حامضية المحلول وسرعة التدوير)‪.‬‬ ‫نتائج التجارب لهذه الدراسة لخصت كالتالي‪:‬‬ ‫تم أستخدام تحليل األرتداد (التراجع) لتعبير رياضي متعدد الحدود من المرتبة الثانية للدالة الهدف‬ ‫(معدل التآكل) ‪ ,‬بأستخدام تصميم فكتوريل الكلي العملي (‪ )FFED‬بواسطة أستخدام البرنامج‬ ‫أألحصائي ‪ , STATISTICA‬أعطى تعبيرين رياضيين لتجارب ذات المتغيرات األربعة‬ ‫والثالثة‪.‬‬ ‫تم أستخدام معادلة أرينيوس (‪ )Arrhenius Equation‬ومعادلة الحالة األنتقالية ( ‪Transition‬‬ ‫)‪ State Equation‬أليجاد قيم طاقات التنشيط ‪ ,‬حيث تم إيجاد الطاقة المنشطة للتفاعل‬ ‫)‪ , (Activation energy, Ea‬أنثالبية التنشيط ( *‪(Enthalpy of activation, △H‬‬ ‫وأنتروبية التنشيط (*‪.)Entropy of activation, △S‬‬ ‫تم أيجاد قييم معيدل ثوابيت أألتيزان )*‪ )K‬عنيد قييم معيدل تغيرطاقية كيبس الحيرة (‪,(△G‬‬ ‫لتحديد تلقائية تفاعل التآكل‪.‬‬ ‫وجد أيضا أن معدالت تآكل الحديد المطاوع في محلول كلوريد الصوديوم بوجود وغياب‬ ‫حامض الخليك تزداد بزيادة درجية الحيرارة ‪ ,‬تركييز حيامض الخلييك (‪ )HAc‬وسيرعة التيدوير‪,‬‬ ‫لكنها تتناقص مع زيادة األس الهيدروجيني (الحامضية) للمحلول‪.‬‬ ‫تم تحليل النتيائج العمليية لتقنيية الفقيدان بيالوزن لمعيدل تآكيل الحدييد المطياوع فيي محليول‬ ‫كلورييد الصييوديوم بوجيود و بغييياب حيامض الخليييك كدالية لدرجيية الحيرارة ‪ ,‬األس الهيييدروجيني‬ ‫(الحامضييية) للمحلييول ‪ ,‬تركيييز حييامض الخليييك (‪ )HAc‬وسييرعة التييدوير وخضييعت لتعبيييريين‬ ‫رياضيين مقترحيين‪.‬‬

‫‪ .2‬طريقة أستقطاب الجهد الحركي ‪(Potentiodynamic Polarization‬‬ ‫)‪Technique‬‬ ‫تم تطبيق الموديل الرياضي المقترح من قبل ‪Korobove and Medvedeva,‬‬ ‫]‪ [2000‬وهذا الموديل يأخذ بنظر األعتبار تأثير أنتقال المادة على التفاعل الكاثودي عن طريق‬ ‫معامل تصحيح ألنتقال الكتلة (‪ )λ‬من تحليل منحنيات أألستقطاب ‪.‬‬

‫‪-2-‬‬



‫السلوك التآكلي للحديد الكاربوني في نواتج المياه‬ ‫للصناعات النفطية المحتوية على غاز ‪CO2‬‬ ‫إعداد‬

‫خالد حامد رشيد عبد الخالق‬ ‫إشراف‬ ‫أألستاذ الدكتور ابرائيل سركيس يارو‬ ‫قسم الهندسة الكيمياوية ‪ -‬كلية الهندسة‬ ‫جامعة بغداد‬ ‫إن وجود غاز ثاني أوكسيد الكاربون ( ‪ (CO2‬في طبقة الماء المالح بشكل غاز مذاب‬ ‫تحت ضغوط عالية والمصاحبة للنفط الخام يكون حامض الكاربونيك‪ .‬الحديد المطاوع يشغل‬ ‫المادة األساسية في تركيب خطوط أألنابيب الناقلة في الصناعة النفطية والغازية ‪ ,‬بسبب رخص‬ ‫ثمنها ‪ ,‬مقاومتها العالية وتوفرها‪ .‬الحديد الكاربوني يتآكل بوجود حامض الكاربونيك والحوامض‬ ‫العضوية مثل حامض الخليك )‪ .(HAc‬لذلك وجب دراسة الظروف التي عندها حامض الخليك‬ ‫يسبب التآكل للحديد المطاوع‪ .‬الظروف المسببة للتآكل بوجود حامض الخليك‪/‬ثاني أوكسيد‬ ‫الكاربون هي ‪ :‬درجة الحرارة ‪ ,‬الضغط الجزئي لثاني أوكسيد الكاربون ‪ ,‬تأثير األس‬ ‫الهيدروجيني (الحامضية) ‪ ,‬نظام الجريان الى آخره ‪.‬‬ ‫إن معدالت التآكل لسبيكة الحديد المطاوع ‪ API X65‬درست بواسطة ثالثة تقنيات‬ ‫مختلفة‪:‬‬ ‫‪.1‬‬

‫تقنية الفقدان بالوزن )‪(Weight Loss Technique‬‬

‫‪.2‬‬

‫تقنية أستقطاب الجهد الحركي ‪(Potentiodynamic Polarization Technique).‬‬

‫‪ .3‬تقنيات تمييز خصائص السطح المتآكل ‪(Characterization of the Corroded‬‬ ‫‪Surface Techniques).‬‬ ‫‪ .1‬تقنية الفقدان بالوزن )‪:(Weight Loss Technique‬‬ ‫أنجزت مجموعة من التجارب لدراسة التأثير للمحاليل المحاكية الملحية على معدل‬ ‫التآكل للحديد المطاوع بوجود وغياب حامض الخليك‪ .‬وجدت نفس معدالت التآكل للحديد‬ ‫المطاوع في المحاليل المحاكية الملحية ومحاليل كلوريد الصوديوم بتركيز ‪( %3.3‬وزنا)‪.‬‬

‫‪-1-‬‬

‫‪CO2‬‬

‫من قبل‬

‫خالد حامد رشيد عبد الخالق‬