A Review of Surfactant As Corrosion Inhibitors

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/318740334
A Review of Surfactants as Corrosion Inhibitors and Associated Modeling
Article in Progress in Materials Science · July 2017

DOI: 10.1016/j.pmatsci.2017.07.006
CITATIONS READS
36 1,234
4 authors, including:
Yakun Zhu Michael L. Free

University of California, Berkeley University of Utah
38 PUBLICATIONS 359 CITATIONS 259 PUBLICATIONS 1,956 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
A novel Ti metal production approach View project
Lithium isotope separation View project
All content following this page was uploaded by Yakun Zhu on 11 February 2018.
The user has requested enhancement of the downloaded file.

Progress in Materials Science 90 (2017) 159–223
Contents lists available at ScienceDirect
Progress in Materials Science

journal homepage: www.elsevier.com/locate/pmatsci
A review of surfactants as corrosion inhibitors and associated

modeling
Yakun Zhu a,⇑, Michael L. Free a,⇑, Richard Woollam b, William Durnie c
a
135 S. 1460 E. Rm 412, Department of Metallurgical Engineering, University of Utah, Salt Lake City, UT 84112, USA
b
BP, 501 Westlake Park Blvd, Houston, TX 77079, USA
c
BP Exploration Operating Company Sunbury, Surrey, United Kingdom
a r t i c l e i n f o a b s t r a c t
Article history: Surfactants have been commonly used as corrosion inhibitors for the protection of metallic
Received 9 March 2017 materials against corrosion. The amphiphilic nature of surfactant molecules creates an
Received in revised form 18 July 2017 affinity for adsorption at interfaces such as metal/metal oxide–water interface. The adsorp-
Accepted 19 July 2017
tion of surfactant on metals and metal oxides creates a barrier that can inhibit corrosion.
Available online 27 July 2017
The properties of surfactant and the interaction of surfactant with metal or metal oxide
and the surrounding solution environments determine the level of adsorption and corro-
Keywords:
sion inhibition. Understanding and modeling the behavior of surfactants in corrosive envi-
Corrosion inhibitors
Interface
ronments is critical to optimal utilization of surfactants as corrosion inhibitors. This review
Adsorption of surfactants as corrosion inhibitors is designed to provide systemic evaluation of various
Aggregation physical and chemical properties of surfactants, surfactant behaviors in corrosive environ-
Partitioning ments, and their influence in corrosion inhibition, which can be used to improve the effec-
Micelle tiveness with which surfactants are used as corrosion inhibitors in a variety of
Modeling studies environments. Progress in the development of various predictive models, including semi-
empirical models, mechanistic models, and multiphysics models, are reviewed for the eval-
uation and prediction of surfactant properties and surfactant corrosion inhibition effi-
ciency. Applications of these models to experimental design and analysis, surfactant
design and selection, and lifetime prediction are also discussed.
Ó 2017 Elsevier Ltd. All rights reserved.
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
2. Corrosion background information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
2.1. Corrosion electrochemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
2.2. CO2 corrosion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
2.3. H2S corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
2.4. Corrosion reaction rate theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
3. Water structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
4. Double layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
5. Mass transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
5.1. Diffusion for short-time start up intervals. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
⇑ Corresponding authors.
E-mail addresses: yakun.zhu@utah.edu, ykzhu1@hotmail.com (Y. Zhu), michael.free@utah.edu (M.L. Free).
http://dx.doi.org/10.1016/j.pmatsci.2017.07.006
0079-6425/Ó 2017 Elsevier Ltd. All rights reserved.
160 Y. Zhu et al. / Progress in Materials Science 90 (2017) 159–223
5.2. Diffusion for slow moving or steady-state systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

5.3. Identifying mass transport control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
5.4. Combination of diffusion and Butler-Volmer kinetics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
6. Surfactant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
6.1. Introduction to surfactant. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
6.2. Hydrophilicity and hydrophobicity of surfactant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
6.3. Adsorption mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
6.4. Surface tension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
6.5. Enhancement of wettability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
6.6. Other properties influenced by surfactants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
6.7. Langmuir-Blodgett films . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
6.8. Krafft point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
6.9. Surfactant states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
6.10. Micelles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
6.11. Microemulsions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
6.12. Surfactant partitioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
6.13. Common surfactant inhibitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
6.14. Surfactant mixtures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
7. Various determining factors of inhibitor efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
7.1. Adsorption at surface/interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
7.1.1. Adsorption isotherms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
7.1.2. Adsorption thermodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
7.1.3. Adsorption kinetics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
7.2. Surfactant aggregation and the aqueous cmc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
7.3. Water/oil partitioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
7.4. Surfactant precipitate and colloid formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
7.5. Fluid flow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
7.6. Salt/ion effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
7.7. Microstructure of metal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
8. Corrosion inhibition experimental evaluations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
9. Corrosion inhibition modeling evaluations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
9.1. Semi-empirical models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
9.2. Mechanistic models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
9.3. Multiphysics models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
9.3.1. General description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
9.3.2. ICI model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
10. Conclusions and perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
Nomenclature
AFM atomic force microscopy

AMT alternative molecular thermodynamic
BAC benzalkonium chlorides
CFU colony forming unit
cmc critical micelle concentration
CVD chemical vapor deposition
DDPB dodecylpyridinium bromide
DFT density functional theory
EDL electrical double layer
FTIR/IRS Fourier Transform Infrared internal reflection spectroscopy
GDP gross domestic product
ICI integrated corrosion inhibition
IHP Inner Helmhotz plane
KV Koroleva and Victorov
LA Langmuir adsorption
LB Langmuir-Blodgett
MF Moreira and Firoozabadi
MI multi-interaction
MLA modified Langmuir adsorption
Y. Zhu et al. / Progress in Materials Science 90 (2017) 159–223 161
MQSAR modified quantitative structure activity relation

MTT molecular thermodynamic theory
NPE nonylphenol ethoxylate
OG octylglucoside
OHP Outer Helmhotz Plane
PB Poisson-Boltzmann
PM-IRRAS polarization modulation infrared reflection absorption spectroscopy
QSAR quantitative structure activity relation
R-H Rubingh and Holland
sac or satc surface aggregation concentration
SA self-assembled
SAED selected area electron diffraction
SDS sodium dodecyl sulfate
TCA trans-cinnamaldehyde
TEM transmission electron microscopy
TS Testing System
WOS water-oil-steel pipe
Symbols
a the activities of the reacting species which can be denoted with appropriate subscripts
Ao optimal surface area of surfactant headgroup
Aa anodic electrode area
Ac cathodic electrode area
Ad area through which the species is transferred in diffusion process
ALB
! surface area of LB film
A0 modified vector of regression coefficients
B0 modified regression constant
C concentration of the reacting species and can be denoted with appropriate super- or subscripts
Cb(a or c) bulk concentration for the anodic (a) or cathodic (c) reactions of the reacting species
Cs(a or c) surface concentration for the anodic (a) or cathodic (c) reactions of the reacting species
Cc concentration of ion dissociated from electrolyte and from ionic surfactant in aqueous solution
Cm overall concentration of total monomers in mixed water-oil environment
Cmo molar concentration of oil
Cmw molar concentration of water
Co total concentration of mixed surfactants in oil, including monomer and micellar form
Cw total concentration of mixed surfactants in water, including monomer and micellar form
C om concentration of total monomers in oil phase
Cw m concentration of total monomers in water phase
C omi concentration of monomeric surfactant ‘i’ in oil phase
Cw mi concentration of monomeric surfactant ‘i’ in water phase
Co surface adsorbate concentration at full coverage
Cs concentration of salt
Ctol initial concentration (not at equilibrium) of total surfactants in water phase
C overall average concentration of total surfactants in water-oil environment
dp diameter of particle
Do standard diffusivity
Df diffusion coefficient of diffusing species in unit cm2s1
e positive elementary charge
e negative elementary charge
E system (or electrode) potential of an electrochemical reaction
Ecorr corrosion potential
ED activation energy for diffusion
Eeq equilibrium potential of a half-cell reaction and can be denoted with appropriate subscripts for different spe-
cies, such as metal M and non-metal X
Eo standard electrochemical potential of an electrochemical reaction and can be denoted with appropriate sub-
scripts for different species
f roughness factor
fi activity coefficient of surfactant ‘i’ in micelles
F Faraday constant
icorr corrosion current density with surfactant inhibitor
iocorr corrosion current density without surfactant inhibitor
ie exchange current density of a half-cell reaction and can be denoted with appropriate subscripts for different
species
io equilibrium exchange current density of a half-cell reaction and can be denoted with appropriate subscripts for
different species
ia anodic current density
ic cathodic current density
ila limiting anodic current density
ilc limiting cathodic current density
Ia anodic current
Ic cathodic current
J flux or rate of mass transfer per unit area
k Boltzmann’s constant
kf adsorption rate constant
kb desorption rate constant
km mass transfer coefficient
00
k a constant that is directly related to the equilibrium exchange current density
K0 modified equilibrium adsorption constant
Kad equilibrium adsorption constant
K haf;1 half saturation constant for interaction 1 (unit less)
K haf;2 half saturation constant for interaction 2 (unit less)
Ki partitioning coefficient of surfactant ‘i’
K ion equilibrium adsorption constant for competing ions
Kmix apparent partitioning coefficient of mixed surfactants
K sur equilibrium adsorption constant for surfactant
lc characteristic value of the length of a tube, or particle diameter, or the length of a plate
lmax the maximum length of surfactant molecule
lm length of the surfactant molecule
ls critical chain length of surfactant molecule
L distance from the leading edge
Li the number of carbon atoms in the hydrocarbon chain of surfactant ‘i’
mi monomer of surfactant ‘i’ (i = 1, 2, or 3. . .)
mj ion ‘‘j” dissociated from salt (j = 1, 2, or 3. . .)
MN micelle with an aggregation number N, micelle composition ai, and a counterion binding coefficient dj
Mad surfactants adsorbed on metal surface and water/oil interface
Mo surfactants distributed in the oil phase
Mtol total quantity of all surfactants added to the WOS environments
Mw surfactants distributed in the water phase
n number of electrons involved in the electrode reaction
nLB number of moles of adsorbed surfactants on LB film
nm number of moles of transferred in diffusion process
N aggregation number of micelle
Nn, Nw, Nz, and NQ the number-based, weight-based, z-based, and quencher-based aggregation number, respectively
Nþ number density for positively charged ions at certain distance from the surface
N number density for positively charged ions at certain distance from the surface
Nþ bulk number density for positively charged ions in bulk solution
N bulk number density for positively charged ions in bulk solution
N cyl
ls
aggregation number per unit length of a cylindrical portion of cylindrical micelle
Nsph aggregation number of a spherical micelle
PLB surface pressure of LB Films
Ps packing parameter of micelle
q charge of the surfactant’s headgroup
r cyl radius of a cylindrical portion of cylindrical micelle
r sph radius of a spherical micelle
rh distance from the location of the charge of headgroup to the surface of the micellar core
R gas constant
Rj reaction rate of ions usually in unit of mol/s per unit area
R Reynolds number
Sad surfactant adsorbed on surface
Saq surfactant in bulk aqueous solution
Sion competing ions
Ssur surfactant adsorbate
Ssn synergistic parameter

t time in unit s
tj transference number
T absolute temperature
uj ion mobility
U
! general bulk solution velocity
U velocity vector
vs a hydrocarbon chain volume
Vo volume of oil phase
Vw volume of water phase
W substrate surface site
xi bulk mixed molar fraction of surfactant ‘i’
X a non-metal species. Xp+ and Xm+ are two different valance states of X
z valence charge of ion and may be denoted with appropriate subscripts for different ions, such as ion ‘‘j”
zi valence of ionic surfactant ‘i’. For nonionic surfactant ‘i’, zi = 0
zj valence of ion ‘‘j”. For nonionic surfactant ‘i’, dj = 0
aa anodic charge transfer coefficients
ac cathodic charge transfer coefficients
ai molar fraction of surfactant ‘i’ in micelle. For micelles of pure surfactant, ai = 1
ba anodic Tafel slope
bc cathodic Tafel slope
c the activity coefficient of the reacting species and can be denoted with appropriate super- or subscripts
cc mean activity coefficient of counterions
comi activity coefficient of monomeric surfactant ‘i’ in oil phase
cwmi activity coefficient of monomeric surfactant ‘i’ in water phase
Cap the average concentration in water-oil environments of mixed surfactants at which mixed micelles start to
form (the apparent cmc of mixed surfactant in water-oil environments)
Co the oil cmc of mixed surfactant
Cw the aqueous cmc of mixed surfactant
0
Cw the initial aqueous cmc of added surfactant mixture before paritioning
ap
Ci the average concentration in water-oil environments of surfactant ‘i’ at which micelles start to form (the appar-
ent cmc for pure surfactant ‘i’ in water-oil environments)
Cpi the aqueous cmc of surfactant ‘i’ in pure water
Cwi the aqueous cmc of pure surfactant ‘i’
di counterion binding coefficient with respect to surfactant ‘i’ based on best-fit of experimental data
dj binding coefficient of ion ‘‘j”
DGoad standard adsorption free energy
D! spreading coefficient
Dloact standard free energy contribution from the activity of surfactant and counterion
Dloelec standard free energy contribution from electrostatic interaction
Dloent standard free energy contribution from entropy gain associated with headgroup-counterion mixing
Dloint standard free energy contribution from formation of micellar core-water interface
Dlom standard micellization free energy per surfactant molecule with consideration of solution composition of sur-
factant and with incorporation of surfactant activity and counterion activity
Dlopack standard free energy contribution from hydrocarbon tail packing in the micelle
Dlost standard free energy contribution from surfactant headgroup steric interaction
Dlotri standard free energy change of transfer of surfactant ‘i’ from aqueous phase to organic phase
Dlotrt standard free energy contributions from hydrocarbon transfer from salt water into micelle
r partial derivatives with respect to the coordinate system
fd thickness of the diffusion layer
fh thickness of hydrodynamic boundary layer
flf thickness of boundary layer for laminar flow over a flat plate
flr thickness of boundary layer for laminar flow over a rotating disc
fls thickness of boundary layer for laminar flow over a spherical particle
ftf thickness of boundary layer for turbulent flow over a flat plate
ga overpotential for anodic reaction
gc overpotential for cathodic reaction
h surface coverage
h1 surface coverages based on surfactant ‘‘1”
h2 surface coverages based on surfactant ‘‘2”
hcalc
1;2 calculated surface coverage of mixed surfactant inhibitors 1 and 2
hmeas
1;2 experimentally measured surface coverage of mixed surfactant inhibitors 1 and 2
H contact angle
H0 contact angle of a rough solid surface
j Debye-length
k(a or c) a factor that depends upon mechanisms of electrochemical reactions
K adsorption density of surfactant and can be denoted with appropriate super- or subscripts
Ke equilibrium amount of adsorption concentration on solid surface
Ki adsorption density of surfactant ‘i’
Kmax;1 maximum adsorption concentration for interaction 1
Kmax;2 maximum adsorption concentration for interaction 2
K adsorption density of ions on solid surface/substrate
m solution viscosity
mk kinematic viscosity
n molecular interaction constant
q solution density
! surface tension (usually with adsorbed layer)
!L surface tension of liquid
!o surface tension without an adsorbed layer
!S surface tension of solid
!SL interfacial tension between solid and liquid
!SO interfacial tension of solid-oil interface
!SW interfacial tension at the solid-water interface
!WO interfacial tension of water-oil interface
rsph surface charge density of spherical micelle
rcyl surface charge density of a cylindrical portion of cylindrical micelle
r surface charge density of solid surface/substrate
1 an empirical fitting parameter in surfactant isotherms
w potential at certain distance from the surface in EDL model
wo surface potential in EDL model
x angular frequency rads1
1. Introduction
This review of surfactants as corrosion inhibitors is designed to provide insights into the processes and mechanisms by
which surfactants inhibit corrosion. General information about corrosion electrochemistry, interfacial water, the electrical
double layer, commonly used classes of surfactants, surfactant behavior, adsorption, mass transport, inhibition mechanisms,
various processes and phenomena that affect surfactant performance in water-oil-steel pipe (WOS) environments, and pro-
gress in various corrosion inhibition models are provided.
2. Corrosion background information
As an important component of the economy, the oil and gas industry has received considerable attention from research-
ers because oil mining and transportation have become increasingly expensive due in part to equipment damage caused by
corrosive media, such as media containing dissolved H2S, Cl, O2, and CO2 [1–5]. As a specific oil and gas industry example,
pipeline made of carbon steel is easily corroded in environments that contain water and carbon dioxide (CO2) [4–6]. The
annual direct cost of corrosion in United State has been estimated to be around $276 billion or 3.1% of the gross domestic
product (GDP). About 3.7% of the total cost comes from the oil and gas industry [3,7,8], much of which is associated with
CO2 corrosion of carbon steel.
Iron, the main constituent of steel, interacts with water, oxygen, and hydrogen ions to form a variety of species. Ideally,
the desired form or species of iron in an application is metallic iron. Iron in steel can react with oxygen from water, or at
elevated temperatures from air, to form iron oxide. If the iron oxide is stable, it can form a layer on the surface of iron that
can slow but not stop further corrosion. Iron oxides react with hydrogen ions to form dissolved iron and water. Hydrogen
ions also directly attack iron. Thus, hydrogen ions are particularly harmful with respect to iron corrosion. Because hydrogen
ion concentration is generally measured using pH (pH is the negative logarithm of the activity of hydrogen ions.), the sta-
bility of iron is a strong function of pH. The relationship between pH and iron species stability is often assessed as a function
of electrochemical potential in Pourbaix diagrams. The electrochemical potential is a measurement of the electron with-
drawing capability of the environment.
Examples of corrosion products on steel surfaces in different aqueous environments are given in Fig. 1 [9]. It is seen that
for the sample polarized in saturated CO2 + 0.01 mol L1 NaNO2 solution there is a continuous and intact layer of corrosion
products with the average thickness about 0.65 mm on the sample surface in Fig. 1(b), whereas the corrosion product layer is
thin and not continuous, as shown in Fig. 1(c), which may be attributed to the insufficient supply of CO2 in the solution. This
is supported by the findings in the corresponding X-ray photoelectron spectroscopy (XPS) spectra in Fig. 1(d) and (e): the Fe
2p spectrum reveals two peaks at 709.60 eV and 711.40 eV, which correspond to Fe element in corrosion products a-FeOOH
and FeCO3, respectively, in the former case (Fig. 1(b)); the Fe 2p spectrum reveals four peaks at 711.52 eV, 710.85 eV,
709.63 eV and 706.10 eV, possibly corresponding to Fe element in corrosion products a-FeOOH, Fe2O3, and FeCO3, and sub-
strate Fe, respectively, in the latter case (Fig. 1(c)). Fig. 1(f) shows atomic force microscopy (AFM) micrographs of the steel
surface after exposure to H2S for 10 s. Fig. 1(g) shows a representative transmission electron microscopy (TEM) image of the
crystalline products and electron diffraction pattern of mackinawite. The selected area electron diffraction (SAED) pattern in
Fig. 1(h) demonstrates the [1 1 1] of mackinawite.
A simplified Pourbaix diagram for iron is shown in Fig. 2 based on thermodynamic data [10,11]. Fig. 2 shows that iron is
stable in its metallic form at low pH if the electrochemical potential is low. The stable region for metallic iron is generally
a b d
0.65 μm
Inner surface
Cross section
c e
f g i
Fig. 1. Corrosion products on steel surface in different aqueous environments: (a) one piece of corroded X65 pipe steel used in oilfields (After Ref. [9]).
Cross-sectional SEM morphologies of mild steel polarized to 1.0 V vs. SCE in (b) CO2 saturated solution containing 0.01 mol L1 NaNO2 and in (c) non-
saturated CO2 + 0.01 mol L1 NaNO2 (After Ref. [4]). (d) and (e) are the high-resolution XPS spectra of Fe 2p corresponding to (b) and (c), respectively (After
Ref. [4]). (f)–(i) shows the corrosion products of mild steel sample surface after exposure to H2S-H2O vapor at 50 °C: (f) AFM image, (g) low-magnification
image showing block-shaped products, (h) SAED pattern showing the [1 1 1] phase of mackinawite, and (i) high-magnification image of mackinawite (After
Ref. [3]).
1.5
Simplified Iron/Water
Pourbaix Diagram
Fe3+
(0.01 m dissolved iron)
1.0
O2
H2O
0.5
Fe2+
0.0 H2O
Fe(OH)3
Water stability
E H2
(V)
-0.5
Fe(OH)2
-1.0
Fe
-1.5
-2 1 4 7 10 13 16
pH
Fig. 2. Simplified (neglects several species) iron/water Pourbaix diagram for 0.01 M of dissolved ferrous ions and 0.01 M of dissolved ferric ions. (After Refs.
[10,11]).
below the level of water stability, which for hydrolysis to form hydrogen is presented as the lower dotted line. Thus, metallic
iron is generally not stable in water. Stable forms of iron in water include Fe2+, Fe3+, Fe(OH)2, and Fe(OH)3 based on the sim-
plified diagram, which neglects some iron species such as Fe3O4 and FeOOH. It is also interesting to note that trihydrated
ferric oxide Fe2O33H2O is stoichiometrically equivalent to Fe(OH)3. Similarly, hydrated ferrous oxide FeOH2O is chemically
related to Fe(OH)2. Consequently, Pourbaix diagrams often show Fe2O33H2O instead of Fe(OH)3, and FeOH2O instead of Fe
(OH)2.
Iron and steel corrosion in aqueous solutions that are not strongly acidic or basic often leads to a combination of oxides
and hydroxides at the surface [12–19]. Often, a magnetite (Fe3O4) layer forms at the metal surface. This layer often restricts
further corrosion. Corrosion can occur through this layer, but it requires charge transfer through it. The conduction of charge
through such a layer utilizes defects in the crystal structure. The outer surface of this magnetite layer is often covered with a
variety of oxides and hydroxides depending on the solution chemistry. This outer layer of oxides and hydroxides is in contact
with the solution and is more porous than the magnetite layer. A schematic diagram of a typical iron or steel surface in aque-
ous media is shown in Fig. 3 [20].
2.1. Corrosion electrochemistry
Corrosion in acidic solutions in oil pipelines containing carbon dioxide and hydrogen sulfide is generally caused by the
reduction of hydrogen as the primary cathodic reaction:
2Hþ þ 2e $ H2 ð1Þ
Net reaction: Fe + 2H+ = Fe2+ + H2
H2O
H+ H+
H2O Fe2+
OH OH +
OH +
2H+ H2
Iron oxides/hydroxides
Fe Fe Fe
Iron oxide barrier layer
Fe
Fig. 3. Illustration of a corrosion scenario involving dilute acid and iron in water.
Other reactions are possible when oxygen availability is significant. The hydrogen ion is generally much more mobile than
dissolved oxygen gas, making the penetration through solution and surface films more viable through direct hydrogen
reduction than through oxygen reduction. Hydrogen and oxygen are readily available from water molecules, and water often
participates in electrochemical reactions when it is split into hydrogen and oxygen components.
One view of iron corrosion in weakly acidic solution is shown in Fig. 4. In typical aqueous solutions iron oxide interacts
with hydrogen and hydroxide ions. These interactions result in surface charging that is pH dependent as indicated in Fig. 4.
At low pH levels, the abundance of hydrogen ions results in the formation of some FeOH+2 species on the surface, creating a
slight positive charge [20]. At intermediate pH levels, the surface tends to remain as FeOH [20]. At high pH levels the surface
has a net negative charge with an excess of FeO at the surface [20]. These charged states influence the adsorption of other
ions on the surface.
2.2. CO2 corrosion
CO2 corrosion in aqueous media, which is often associated with pitting corrosion, is an electrochemical process involving
electron transfer that leads to the dissolution of iron at the anode and the evolution of hydrogen at the cathode. Corrosion
products within pits mainly consist of FeCO3 [21–24], regardless of the surrounding CO2 pressure [25]. Upon local anodic
iron dissolution a corrosion product layer of FeCO3 grows on the pit surface due to its low solubility in water (solubility pro-
duct is 10.54 at 25 °C) [23,26,27]. The overall reaction is:
Fe þ CO2 ðgÞ þ H2 OðlÞ $ FeCO3 þ H2 OðgÞ ð2Þ
In acidic solution, the anodic reaction follows a general multi-step mechanism [21–27]:
Fe $ Fe2þ þ 2e ð3Þ
Fe2þ þ CO2
3 $ FeCO3 ð4Þ
Fe2þ þ 2HCO3 $ FeðHCO3 Þ2 ð5Þ
FeðHCO3 Þ2 $ FeCO3 þ CO2 þ H2 O ð6Þ
at the cathode,
CO2 þ H2 O $ H2 CO3 ð7Þ
2HCO3 þ 2e $ H2 þ 2CO2

3 ð8Þ
HCO3 $ Hþ þ CO2
3 ð9Þ
Based on these corrosion processes (Eqs. (3)–(9)) under appropriate conditions, a corrosion scale FeCO3 is expected to
deposit on the surface of the carbon steel.
There are many factors that affect the anodic dissolution of iron in CO2-saturated solution. A change in temperature can
lead to a change in CO2 corrosion [25,28–31] and chemical composition of corrosion product [32,33]; an increase in CO2 par-
tial pressure [21,28] and fluid flow rate [22,34,35] accelerated CO2 corrosion. An increase in pH [21,25,32] and solution salt
concentration such as Cl [36] and Cr [37] content in the steel matrix can decrease the uniform corrosion rate. Microstruc-
1
0.9
0.8
0.7
Fracon of Species
0.6 FeOH2+
0.5 FeOH
0.4 FeO-
0.3
0.2
0.1
0
0 2 4 6 8 10 12 14
pH
Fig. 4. Comparison of pH and surface speciation on iron based on data in Ref. [20].
tural features, such as scratches and microcracking, promote corrosion processes [23,29]. In addition, O2 can enhance CO2
corrosion rate by acting as a catalyst [38] and H2S can increase corrosion through its synergistic action with CO2 [39].
For a surface with microcracking due to the characteristic microstructure such as high energy phase/grain boundaries,
carbide precipitation, and coarsened dislocations, etc., it is reported that the formation of the FeCO3 scale in geothermal
water generally takes place in two reaction steps: (1) the Fe2+ compounds in which ferrous hydroxide Fe(OH)2 film forms
in the first step as shown in Eq. (10) [23,27], leading to an increase of the local pH. (2) At the saturation dissolution of ferrous
hydroxide a Fe(OH)2-passivating film can form, followed by the precipitation of a ferrous carbonate scale FeCO3 (see Eqs. (11)
and (12)) due to the low solubility of FeCO3 in water [23,40,41]. Growth of the corrosion layer depends on the carbon and
oxygen partial pressures. If the ferrous carbonate film is highly porous it may initiate local depassivation of the steel, leaving
the activated surface susceptible to local corrosion induced microcracking [23].
Fe þ 2H2 O $ ½FeðOHÞ2 ads þ 2Hþ þ 2e ð10Þ
½FeðOHÞ2 ads þ ½H2 CO3 ads $ FeCO3 þ 2H2 O ð11Þ
Fe þ CO2
3 $ FeCO3 þ 2e ð12Þ
The characteristics and morphology of corrosion product film FeCO3 are heavily dependent on temperature which, in
turn, influences the CO2 corrosion process [25,28–31]. At modest temperatures (<70 °C), anodic dissolution of iron progres-
sively increases with temperature [42]. It is believed that the increase of corrosion rate in the low-temperature range is
attributed to an increase of mass-transfer rate due to flow effects and slow FeCO3 formation process [29,42]. The corrosion
rate gradually diminishes with FeCO3 growth. However, the corrosion process proceeds unhindered at sites of FeCO3 break-
down, which may lead to severe localized attack [42]. At temperatures above 80 °C, the solubility of FeCO3 in aqueous solu-
tion decreases, which eventually leads to FeCO3 precipitation in solution and coverage on steel surfaces [29,42]. Therefore, a
diffusion process may become the rate-determining step in CO2 corrosion of iron after the formation of the protective scale.
On the other hand, the above discussion does not rule out the possibility of the occurrence of localized corrosion or the for-
mation of nonprotective corrosion scales because the diffusion barrier and the porous layer FeCO3 may be superseded by the
liquid surface states [29].
Geological formation water in oil and gas wells usually contains dissolved salts at high concentration and it is necessary
to investigate the CO2 corrosion mechanism at high salt concentrations to simulate the commercial environment. It has been
found that the corrosion rate of iron is inhibited with the increase of Cl concentration [37,40,43] probably because of the
decreased solubility of CO2 in aqueous media. However, it is also reported that Cl content is important in the onset of local-
ized corrosion, and the presence of a small amount of Cl could significantly reduce the passivating tendency of steel due to
the increased ionic strength and solution conductivity [44,45]. Very recently it has been reported that the maximum iron
corrosion rate (around 86 lA cm2) in simulated CO2-saturated oil well environments is reached at Cl (or NaCl) concentra-
tion of 25 g/L [36]. Below the maximum corrosion rate, CO2 corrosion is promoted with increasing Cl concentration. Above
the maximum corrosion rate, however, CO2 corrosion is inhibited with increasing Cl concentration because of the signifi-
cantly reduced CO2 solubility and the decreased opportunities of H+, H2O, H2CO3 and HCO 3 to participate in corrosion. More-
over, Cl can destroy the corrosion product films and change the morphology of corrosion product films, which are in line
with previous findings [46,47]. Despite these studies, research on the effect of Cl over wide concentration ranges on CO2
corrosion is limited. The catalytic mechanism of Cl in promoting anodic dissolution is summarized as follows [36,48,49]:

Fe þ Cl þ H2 O $ ½FeClðOHÞ ads þ Hþ þ e ð13Þ
½FeClðOHÞ ads $ FeClOH þ e ð14Þ

FeClOH þ Hþ $ Fe2þ þ Cl þ H2 O ð15Þ

The number of holes and cracks in the corrosion product films increases as the Cl content increases, and the corrosive
medium can pass through these areas to penetrate the films. Consequently, the corrosion of the metallic matrix is acceler-
ated [36].
2.3. H2S corrosion
Hydrogen sulfide H2S interacts with the surface to form complexes with iron and provide hydrogen ions. Iron dissolution
in aqueous solutions containing H2S is often believed to occur as follows [50–54]:
Fe þ H2 S $ FeHS þ Hþ ð16Þ
FeHS $ FeHS þ e ð17Þ
FeHS $ FeHSþ þ e ð18Þ

FeHSþ $ FeS þ Hþ ð19Þ

The formation of iron sulfide layers generally results in films that may be incomplete and/or contain many defects, mak-
ing it easier to corrode iron with iron sulfide than surfaces that contain iron oxide or iron oxide hydroxide. In addition,
because iron sulfide has some conductivity and a potential that can be higher than that of iron in some environments, iron
sulfides can form detrimental galvanic couples in steel. The sulfur availability can lead to the formation of iron sulfide com-
pounds such as mackinawite.
Corrosion of steel is further complicated by the presence of additional phases such as cementite (Fe3C) which result in
microscopic galvanic couples that are well distributed at the surface. The presence of these cementite inclusions disrupts
films, such as iron oxide layers, and accelerates local corrosion. Thus, cementite increases corrosion and reduces oxide film
integrity.
Thus, there are many reactions that are possible at a steel surface in solutions containing carbon dioxide and hydrogen
sulfide that are common in oil wells, which are also commonly acidic and contain dissolved salts such as sodium chloride.
2.4. Corrosion reaction rate theory
Corrosion is driven by electrochemical reactions. For an electrochemical reaction given as:

X pþ þ ne $ X mþ ð20Þ
p+ m+
where X represents a non-metal species, X and X are two different valance states of X, e is electron charge, n is the num-
ber of electrons involved in the electrode reaction. Note that similar reaction is also applicable to metal species M.
The resulting expression for the rate of reaction in terms of exchange current density is [55]:

aa FðE Eeq Þ ac FðE Eeq Þ
ie ¼ io exp exp ð21Þ
RT RT
This is a common form of the Butler-Volmer equation. Current density usually has units of A cm2 or lA cm2. The sub-
script ‘‘a” denotes anodic, the subscript ‘‘c” denotes cathodic. The collective term ‘‘E Eeq ” is often called the overpotential in
units of mV or V. It represents the potential ‘‘over” the equilibrium potential for a specific half-cell reaction. aa and ac are
anodic and cathodic charge transfer coefficients, respectively. Note that this equilibrium potential is not the same as the
reaction or corrosion potential, which results from the interaction of two or more half-cell reactions. A negative overpoten-
tial indicates a potential below the equilibrium potential. If the system (or electrode) potential, E, is above equilibrium
potential Eeq, the first (anodic) exponential term in the Butler-Volmer equation dominates. If the corrosion potential is below
Eeq, the second (cathodic) term dominates. In other words, if the corrosion potential is below Eeq, the current flow is negative
or cathodic. Negative currents indicate a net reduction reaction. Positive currents indicate net anodic current. io in the equa-
tion is the equilibrium exchange current density. F is the Faraday constant. R is the gas constant. T is the absolute
temperature.
The equilibrium potential is determined using the Nernst Equation:
RT aXmþ 0:0591 cXmþ C Xmþ
Eeq ¼ Eo ln ¼ Eo ln ð22Þ
zF aXpþ z cXpþ C Xpþ
Eo is the standard electrochemical potential for the reaction. The activities of the reacting species are denoted by ‘‘a” with
appropriate subscripts. The concentration C is given in terms of molality. The activity coefficient is expressed as c and is
dimensionless.
A modified form of the Butler-Volmer equation that is explicit in terms of surface and bulk solution concentrations that
contribute to the exchange current density can be expressed as [56]:
" ka kc #
00 C sa aa FðE Eeq Þ C sc ac FðE Eeq Þ
i ¼ k C ba exp C bc exp ð23Þ
C ba RT C bc RT
00
where k is a constant that is directly related to the equilibrium exchange current density, Cb(a or c) is the bulk concentration
for the anodic (a) or cathodic (c) reactions, Cs(a or c) is the surface concentration for the anodic (a) or cathodic (c) reactions, and
k(a or c) is a factor that depends upon reaction mechanisms and related factors, and it is usually between 0.25 and 1. The con-
centration at the surface is generally not the same as the concentration in the bulk solution.
The Butler-Volmer equation provides the current density for one half-cell reaction. Complete corrosion reactions involve
at least two coupled half-cell reactions. Coupled half-cell reactions must have balanced currents. In other words, the cathodic
current Ic must balance the anodic current Ia . If more than two reactions are coupled, the sum of currents must balance.
Mathematically, the current balance can be expressed as:
X X X X
Ic ¼ ic Ac ¼ ia Aa ¼ Ia ð24Þ
where Aa and Ac are anodic electrode area and cathodic electrode, respectively. ia and ic are anodic and cathodic current den-
sities, respectively. If the area of reaction is the same for all reactions, current densities can replace currents. A more com-
plete discussion of the use of the Butler-Volmer equation and associated applications is discussed elsewhere [57,58].
The mixed or corrosion potential is often far from the anodic and cathodic reaction equilibrium potentials. An example is
shown in Fig. 5 [59]. The example shows that at the corrosion potential the potential is linearly related to the absolute value
of the logarithm of current density. The linearity near the corrosion potential is due to the dominance of one term in the
Butler-Volmer equation. Consequently, the respective Butler-Volmer expressions can be simplified to one exponential term
when analyzed far from equilibrium. The linear region is usually well-established approximately 150–200 mV from the equi-
librium potentials for typical reactions. If the individual half-cell reaction potentials are near the corrosion potential, the lin-
earity will not be present in the potential versus current density plotting format.
In Fig. 5, M represents a metal and X represents a non-metal species. The intersection of the potential versus current den-
sity lines for M and X occurs at the corrosion or mixed potential. In electrochemical literature the potential at the intersec-
tion of M and X lines is also known as the reaction potential. Because it is based on the mixture of two or more reactions, it is
a mixed potential, and for corrosion systems it is known as the corrosion potential. M and X reaction rates at the corrosion
potential are equivalent in magnitude but opposite in sign. Thus, the respective anodic and cathodic terms are

aa FðE EeqM Þ
iM ¼ ioM exp ð25Þ
RT

ac FðE EeqX Þ
iX ¼ ioX exp ð26Þ
RT
in which the subscript ‘‘a” denotes anodic and the subscript ‘‘c” denotes cathodic. The cathodic reduction rate iX for non-
metal X and the anodic metal oxidation rate iM for metal M are equal to the system corrosion/reaction rate iocorr for a two
reaction system. The rate of a reaction can be determined by the reaction rate at the mixed or corrosion potential. Rearrange-
ment and the use of the logarithm function for this specialized case of the modified Butler-Volmer Equation, which is only
applicable when the potential is about 200 mV or more from the equilibrium potential for the reaction, leads to:

RT iM
E EeqM ¼ ln ð27Þ
aa F ioM
Substitution and further rearrangement for this anodic expression leads to:

2:303RT iM
ga ¼ log10 ð28Þ
aa F ioM
Rearrangement and substitution lead to the form known as the Tafel equation:

iM
ga ¼ ba log10 ð29Þ
ioM
In which the anodic Tafel slope, ba , is given by:
2:303RT
ba ¼ ð30Þ
aa F
Fig. 5. Qualitative comparison of potential and current density (log10-scale) for hypothetical reactions. (After Ref. [59]). This plot is based on Butler-Volmer
electrochemical reaction kinetics.
The analogous expressions for the cathodic reaction are:

jiM j
gc ¼ bc log10 ð31Þ
jioM j
and
2:303RT
bc ¼ ð32Þ
ac F
These equations show that the logarithm of the current density is directly proportional to overpotential. The latter forms,
which include overpotentials, ga and gc , are known as Tafel equations. The slopes of the resulting lines are ba and bc. These
slopes are known as Tafel slopes. The associated lines, which are dashed in Fig. 5 are Tafel lines. The Tafel lines are often used
in place of the Butler-Volmer lines to simplify plots. The Tafel lines are only appropriate when the analysis is performed far
from equilibrium (often >150–200 mV from Eeq for each reaction) and the rate is controlled by the electrochemical reactions,
rather than by mass transport.
The corrosion potential and corrosion rate are often influenced by mass transport. The reaction at a corroding surface
must be supplied with the oxidant ions needed for the reaction. The transportation of ions to the reacting interface often
limits the rate of the reaction. Consequently, under mass transport limited conditions, the rate is independent of potential.
Mass transport limitations can be increased by additional convective mass transport, such as can be imparted by stirring as
illustrated in Fig. 6.
3. Water structure
The structure of water at metal and metal oxide surfaces has been studied extensively by many researchers. Most studies
under high vacuum or aqueous conditions reveal that water attaches to pure, close-packed, noble metal surfaces with weak
interactions [60–63]. In the case of Pt (1 1 1) and Ru (0 0 1) some ordered domains are observed at low temperatures [63–
65]. However, when oxygen is added to the surface, water (or deuterium oxide, which is often used in infrared studies) mole-
cules interact strongly with oxygen and form ordered structures [66,67]. The adsorption of water on metal surfaces with
added oxygen is sufficiently strong to form monomer layers with low water molecule mobility [60]. In contrast, on pure
metal surfaces without oxygen the adsorption often leads to clusters, hexamers [63] and high mobility of adsorbed water
molecules [60].
Most of the high precision measurements of water molecules and their behavior on metal surfaces have been performed
at very low temperatures (20–140 K). Consequently, the associated results may not have direct relevance to corrosion at
300 K. However, a number of studies have also been performed under ambient conditions.
Water is often bound tightly to hydrophilic surfaces. However, it should be noted that the water molecules at the surface
are actively moving around on the surface. Adsorbed water molecules are far less mobile on a hydrophilic surface than they
are in the bulk water. Water molecules near the surface tend to become structured in a pattern that is complimentary to the
water molecules on the surface. Subsequent water molecules tend to alternate in layers with water molecules oriented in a
complimentary way to nearby water molecules and layers. It has been observed that as two hard, smooth surfaces approach
each other within about 1 nm, the force between the two objects changes with a spatial periodicity that corresponds to the
iocorr (low flow)
iocorr (high flow)
Fig. 6. Qualitative comparison of potential and the logarithm of the absolute value of current density for metal ‘‘M” and oxidant ‘‘X” under two cathodic
mass transport limited fluid flow conditions. (After Ref. [59]).
size of individual water layers or about 0.25 nm [68]. In other words, as two hydrophilic surfaces approach each other the
force needed to bring them into contact changes rapidly in an oscillating pattern with distance as the surfaces come within
1 nm of each other. These results indicate water molecules can behave as if some oriented water molecules exist near hydro-
philic surfaces [68].
This force associated with water has been referred to in connection with hydration forces, and hydration forces have been
used or misused to explain many surface phenomena. There is some evidence that the steep short range repulsive forces may
be due to entropic repulsion arising from confinement of mobile surface groups [69]. Efforts have been made to use salts that
either enhance the observed interaction forces (‘‘structure makers”) or diminish the forces (‘‘structure breakers”) [69]. How-
ever, although water orients in effective layers near the surface, it is only the first layer of water that binds to the surface, and
subsequent layers interact with each other with the same binding force as is found in bulk water. Therefore, the properties of
water away from the initial layer are similar to those of bulk water, although the interactions of the initial water layer lead to
some orientation of subsequent layers near the surface. Moreover, the use of ‘‘structure breakers” or ‘‘structure maker” salts
should be confined to their interactions with surfaces and the associated effect on the first layer of water rather than on the
modification of bulk water.
4. Double layer
Electrified interfaces cause ions and water molecules to align at surfaces. At positively charged surfaces, negatively
charged ions tend to form a layer very near the positively charged surface. Water molecules align as oriented quadrapoles
(oxygen aligned near the positively charged surface with its two electron rich lobes and the two hydrogen atoms with their
positive rich regions near the negatively charged anions) between the charged surface and the associated layer of counter
ions. Similarly, a negatively charged surface attracts a layer of positively charged cations as well as appropriately oriented
water molecules. The result of a charged surface and a layer of oppositely charged ions separated by the dielectric water
molecules is a capacitor. Because there are two layers of charge involved, it is a double layer. The double layer is usually
called the electrical double layer (EDL). A simple model of a double layer and associated water molecules and ions is shown
in Fig. 7. There is a very large electrical potential gradient that forms between the surface and the edge of the first layer of
counter ions. A general model for the simple double layer, the Helmholtz model, which is applicable to concentrated elec-
trolytes assumes the potential decays linearly between the surface and the first layer of adsorbed, hydrated ions, which have
a charge opposite to that of the surface as shown in Fig. 7. The first layer of adsorbed hydrated ions with opposite charge is
known as the Outer Helmhotz Plane (OHP). The OHP marks the plane at which the double layer capacitance is present with
the first and second water layers forming the dielectric medium. In some cases the OHP is also known as the Stern plane. The
Inner Helmhotz plane (IHP) is the plane at which contact adsorbed ions are located. There are some variations to the general
view shown in Fig. 7, although the variations are generally minor.
Outward from the surface beyond the first layer of counter ions, other ions are found with charges opposite to the first
layer. These ions tend to associate near the first layer, but the organization of layers of alternating charge is not generally
present. Thus, beyond the first layer of charged ions adsorbed at the surface there is a diffuse charge region as illustrated
in Fig. 7. The overall view shown in Fig. 7 is closely related to the Stern model, although the Stern model was based more
closely to a capacitive view of the interface [51].
A common model for the diffuse electrical double layer, which applies beyond the OHP, is the Gouy-Chapman model. The
Gouy-Chapman model for potential as a function of distance from a surface in the diffuse region, which is generally very
useful for dilute solutions, is [70]:
" e#
kT 1 þ ejx tanh zw o
w¼ ln zw
4kT
ð33Þ
2ze 1 ejx tanh 4kTo e
The associated distribution of positive and negative ions as a function of distance from the surface in the diffuse region is
[71]:

zew
Nþ ¼ Nþbulk exp ð34Þ
kT

zew
N ¼ Nbulk exp ð35Þ
kT
N þ and N is the number density for positively and negatively charged ions as noted by the signs of the superscripts at the
potential specified based on the bulk values denoted by the subscript. z is the charge of the ion, e is the charge of an electron,
w is the potential, wo is the surface potential, k is Boltzmann’s constant, and T is absolute temperature. j1 is the Debye-
length. The potential between the surface and the OHP is determined based on linear interpolation using a straight line.
The surface potential is determined by the applied potential or by the potential determining ions, which are those ions
that are adsorbed more readily than other available ions. Often, hydrogen and hydroxide ions are potential determining ions.
Consequently, surface potential is generally a strong function of pH. The pH (or concentration of other potential determining
ions) at which the surface is neutral in the absence of an applied potential is known as the point of zero charge [72].
Fig. 7. Schematic diagram of electrical double layer and the related potential versus distance relationship for potential for a sample electrolyte showing ions
and oriented water molecules, which are shown as circles with arrows to illustrate positive orientation towards partially charged hydrogen atoms. (After
Ref. [55]).
The actual surface potential is not measured directly. Instead, the Zeta potential, which is the effective potential observed
by charged particle motion in an electric field, is measured. It is often assumed that the OHP and the shear plane are the same
for purposes of extrapolating to estimate the surface potential. The pH at which the surface has no net charge is known as the
isoelectric point (IEP). The charge on surfaces at which no external electrical field is applied often influences the adsorption
of other ions such as surfactants. Most iron oxide minerals such as magnetite and hematite have IEP values of 6.7, and tend to
have a positive charge in acidic media [72]. However, as the ionic strength or the dissolved solids concentration increases,
the effect of the surface charge become less influential, and the adsorption of similarly charged ions is often feasible.
In addition, it should be noted that the adsorption of large molecules at an electrode surface modifies the electrical double
layer significantly. In particular, adsorbed molecules move the shear plane, also known as the slip plane, making it less likely
to be near the OHP. A typical adsorbed surfactant molecule is around 2 nm long. Thus, the slip plane may be moved from the
OHP, which is often a few angstroms from the surface, to 2 nm.
5. Mass transport
Mass transport in solution precedes corrosion and adsorption. The relationship between mass transport and corrosion is
very important because many corrosion reaction rates are controlled by mass transport. Consequently, corrosion related
mass transport issues have been studied and published [73,74].
Mass transport of ions in aqueous media often involves three processes. These processes are convection, diffusion, and
migration. Convection is simply bulk mass transport of ions by fluid flow. Diffusion is mass transport involving molecular
movement of species in response to a concentration gradient. Migration is mass transport in response to an electrical poten-
tial gradient. Migration completes the flow of charge in the electrical circuit. These processes are also associated with reac-
tions. The rates of these processes must be considered in a mass balance. Thus, a general mass balance for ion species ‘‘j” in a
small element of fluid leads to [56]:
@C j !
¼ U rC j þ zj F rðuj C j rEÞ þ rðDf rC j Þ þ Rj ð36Þ
@t
C j is the concentration of ion ‘‘j”, Df the diffusion coefficient of diffusing species in unit cm2 s1, E is electrical field, Rj is reac-
!
tion rate usually in unit of mol/s per unit area, t is time in unit s, U is the velocity vector, uj is ion mobility (usually in unit
cm2 V1 s1), and zj represents ion charge. The gradient operator, r, represents the partial derivatives with respect to the
coordinate system. Other symbols are the same as defined previously. The first term on the right-hand side of the equation
accounts for convection. The second term on the right represents migration. The third term on the right is for diffusion. The
last term represents the reaction rate. The term on the left-hand side of the equation provides the rate of change of concen-
tration or the overall change in concentration with respect to time. In most applications, one or more of the terms can be
neglected. In steady-state application scenarios the left-hand side of the equation is zero.
The processes of convection, diffusion, and migration are illustrated in Fig. 8 [77]. Fig. 8 shows an electrochemical cell that
produces hydrogen and bromine from hydrobromic acid. It illustrates the quantitative dimensions for ion transport mecha-
nisms. In the case of hydrobromic acid, hydrogen ions are 4 times more mobile than bromide ions. Consequently, the hydrogen
ions transport more net charge than bromide ions. The fraction of the migration current flux carried by a specific ion is propor-
tional to its concentration and mobility relative to that of the system, and it is calculated by the transference number tj [77]:
zj Fðzj uj FC j rEÞ z2j uj C j
tj ¼ 2P 2
¼P 2 ð37Þ
ðF j zj uj C j ÞrE j zj uj C j
The migration of bromide and hydrogen ions is necessary to complete the flow of charge in the circuit. Migration is based
on ion availability and mobility as indicated in the mass balance equation. Migration is generally not significant in corrosion
contexts, but it can be under unusual circumstances, such as in some systems where cathodic protection is applied.
Diffusion is needed to provide the ions not transported by migration. Diffusion is also needed to remove ions used for
migration that are not consumed by reaction. In the case of a hydrobromic acid hydrolysis cell (see Fig. 8), diffusion provides
one fifth of the hydrogen ions and four fifths of the bromide ions needed.
Ions are transported through the bulk medium by convection. Convection can occur naturally by thermal gradients or it
can be forced by mechanically induced fluid flow such as by pumping or stirring. Convection is needed to transport ions near
the surface where they can diffuse to the surface. However, diffusion is always needed to transport at least some of the
needed ions to the reacting surface. In contrast, forced convection is not necessary for corrosion to occur. Increased convec-
tion also thins the diffusion or mass transport boundary layer thickness, thereby decreasing the diffusion distance and
increasing the flux of oxidant to the corroding surface.
Most applications allow for simplification of the general mass balance expression. The migration term can be eliminated
in the absence of an electrical field. Migration can also be neglected when the background electrolyte concentration is high
relative to the active species concentration. Interfacial reactions often incorporate the effects of convection in a mass trans-
port boundary layer and eliminate the convection term. The reaction term is often equated to the change in concentration
with respect to time. If these assumptions and simplifications are appropriate, the equation can be simplified to:
Fig. 8. Ion transport illustration for an electrochemical cell. Hydrobromic acid is used as the example medium. (After Ref. [77]).
@C j
¼ rðDf rC j Þ ð38Þ
@t
This equation is also known as Fick’s second law. Other simplifications lead to other forms of the equation.
5.1. Diffusion for short-time start up intervals
If the reaction occurs rapidly, the thickness of the reaction zone becomes very small and the reaction interface moves
rapidly. In such cases the reaction position is moving at a rate similar to the diffusion rate. The appropriate expression
for this application is Fick’s second law, which for constant effective diffusivity can be written as [27]:
@C j
¼ Df r2 C j ð39Þ
@t
where r2 has different expressions in different coordinate systems selected for the problem.
5.2. Diffusion for slow moving or steady-state systems
Interfacial adsorption of inhibitors requires mass transport in the form of diffusion. If systems do not involve migration
and are nearly steady-state and slow moving, the flux is approximated by Fick’s first law:
dnm 1 dC j
J¼ ¼ Df ð40Þ
dt Ad dfd
Ad is the area through which the species is transferred. J is the flux or rate of mass transfer per unit area. nm is the number of
moles of transferred. fd is the thickness of the diffusion layer. fd may be in terms of radius or other spatial coordinates
depending upon the coordinate system. The negative sign indicates the direction of transport toward the surface relative
to the direction vector. This form of Fick’s Law only applies to reactions for which the reaction front can be considered as
a stationary front relative to diffusion. It is often applicable to mineral leaching and metal corrosion. Application of this
approach to particle leaching requires discrete particle sizes [75]. The same solution method can be applied to corrosion
through porous layers. It can also be used for other aqueous reactions involving slow moving reaction interfaces [58].
The thickness of the diffusion layer, fd , depends on the conditions. Reactions involving complete dissolution require dif-
fusion through a solution boundary layer. This boundary layer is the layer of medium through which ions must be trans-
ported by diffusion rather than by convection. However, it is often significantly different than the hydrodynamic or
momentum-based, fluid flow boundary layer thickness. The ratio of the hydrodynamic boundary layer thickness to that
of the diffusion boundary layer thickness can be estimated using the equation [76]:
1=3
fh m
¼ ð41Þ
fd qDf
In this equation, fh is hydrodynamic boundary layer thickness, m is the solution viscosity, and q is the solution density.
Using typical solution values of 0.01 g/(cm s) for viscosity, 1 g/cm3 for density, and 1 105 cm2/s for ions diffusion coeffi-
cient in water leads to a ratio of 10 based on this equation. In other words, the hydrodynamic boundary layer thickness is
approximately 10 times greater than that of the diffusion boundary layer.
As the distance from the surface increases the solvent molecules become more mobile. Solution in the bulk is usually
moving due to natural or forced convection. Consequently, movement of bulk fluid transports ions to other zones in a sys-
tem. However, bulk fluid flow does not transport ions into regions very close to the surface. Thus, the boundary or diffusion
layer represents an approximate boundary between the transport of ions by diffusion and convection. Figs. 9 and 10 illus-
trate the interactions among the reaction interface, convection, and diffusion in relation to the diffusion boundary layer.
The thickness of the boundary layer takes on different forms depending upon the geometry and flow as indicated by the
equations in the next few paragraphs [78–80]:
The boundary layer thickness for laminar flow over a flat plate is:
U 1=2 mk1=6 D1=3

1=2
flf ¼ 3l f ð42Þ
where l is distance from the leading edge. U is the bulk solution velocity. mk is the kinematic viscosity.
The boundary layer thickness for turbulent flow over a flat plate is:
U 9=10 mk17=30 D1=3

1=10
ftf ¼ l f ð43Þ
The boundary layer thickness for laminar flow over a rotating disc is:
flr ¼ 1:6x1=2 m1=6

k Df
1=3
ð44Þ
where x is the angular velocity.

Fig. 9. Schematic illustration of a reactive ion that will participate in a reaction at the metal. (After Ref. [58]).
Fig. 10. Schematic illustration of fluid flow and ion transport near a reaction interface. (After Ref. [54]).
Laminar flow around a spherical particle gives a boundary layer thickness (based on the Frossling correlation [81]):
dp
fls ¼ ð45Þ
2 þ 0:6m1=6 dp U 1=2 Df1=3
1=2
k
where dp is the diameter of the particle.

Laminar and turbulent flow conditions must be determined prior to use of these equations. The evaluation is performed
by first calculating the Reynolds number, R:
qUlc Ulc
R¼ ¼ ð46Þ
m mk
where U is the characteristic velocity or generally the bulk velocity. lc is the characteristic value of the length of a tube, or
particle diameter, or the length of a plate. The Reynolds number is used to determine the appropriate model. The Reynolds
number is the ratio of inertial to viscous force. High Reynolds numbers indicate turbulent flow. Low Reynold numbers are
associated with laminar flow. The transition between laminar and turbulent flow in pipes occurs at R 2300 [76]. For flow
around particles the transition begins above a Reynolds number between 0.1 and 1 [76].
Mass transport is sometimes calculated using a mass transfer coefficient. In some texts, Fick’s first law is expressed in a
rearranged form:
dnm
¼ km Ad ðC b C s Þ ð47Þ
dt
km is the mass transfer coefficient and equal to diffusivity (diffusion coefficient) over boundary layer thickness. Conse-
quently, mass transfer coefficients are used in many applications. Use of mass transfer coefficients requires compliance with
Fick’s first law. Diffusivity and/or equivalent boundary layer thickness values can be obtained from mass transfer
coefficients.
The diffusivities of various species in water depend on conditions. Oxygen’s diffusivity is 2.25 109 m2/s in pure water
[82]. However, in more typical solutions, oxygen’s diffusivity is 0.5 109 m2/s [81]. The diffusivity for dissolved chlorine
gas (in solutions with 0.12 m of dissolved salt) is 1.26 109 m2/s. Diffusivity for chloride ion in sodium chloride
(0.05 m) is 1.26 109 m2/s at room temperature [83]. Diffusivity data are commonly found in many texts. The Stokes-
Einstein equation can be used to estimate diffusivity for small, nearly spherical ions or molecules [76]:
kT
Df ¼ ð48Þ
4mpr
where m is the solution viscosity (note that viscosity is a strong function of temperature also). r is the effective radius of dis-
cussed species. When using the Stokes-Einstein equation, some caution should be exercised. The radius of the ions, especially
multivalent ions, may change with temperature. The radius change is often associated with hydration number. The Stokes-
Einstein equation is often not accurate. However, it can be used to predict diffusivity changes due to changing viscosity and
radii.
Another approach to determine diffusivity invokes molecular motion theory. Diffusion is a molecular motion process that
is similar to a chemical reaction. Reactions and diffusion involve internal energy and the probability of getting to the next
step or position. In the case of reaction, the next step is usually interaction with another particle or atom. In the case of dif-
fusion, the next step is physical displacement or movement. A simplified expression for diffusion that utilizes molecular
motion theory is [76]:

ED
Df ¼ Do exp ð49Þ
RT
In this expression Do is the standard diffusivity. ED is the activation energy for diffusion.
5.3. Identifying mass transport control
Mass transport control can be identified by appropriate measurements and analysis. If increasing fluid velocity increases
reaction rate, mass transport participates in kinetic control. If a fluid velocity increases and no change in rate is observed, the
reaction is not controlled by mass transport to the outer surface. It may still be controlled by inner diffusion through a cor-
rosion product layer or a coating if increased fluid flow does not impact the rate. If the increase in fluid velocity causes the
rate to increase in proportion to mass transport control equations, the rate is mass transport controlled.
Using Fick’s first law, the rate of mass transport controlled reactions is inversely proportional to boundary layer thickness.
Thus, if boundary layer thickness is decreased by 50%, the rate doubles. The boundary layer thickness is inversely propor-
tional to the square root of rotational velocity for rotating disc experiments. Thus, for rotating disk experiments the rate
should double if the rotational velocity is increased by four times if it is mass transport controlled. The data in Fig. 11 shows
a comparison of the square root of the rotational velocity to the reaction rate for a rotating disk experiment. The plot sug-
gests the reaction is mass transport controlled because the rate increases in proportion to the square root of rotational
velocity.
A nonlinear relationship between rate and square root of rotational velocity in Fig. 11 would suggest mass transport and
reaction control. A horizontal line would indicate reaction or inner diffusion control. Similarly, the boundary layer thickness
equations can be used to compare observed rate changes to the rates expected for mass transport control. If the rate change
is not the same as expected, the rate may be controlled by both reaction and mass transport kinetics. Often, the overall rate is
controlled by both mass transport and reaction kinetics. In such cases a combined model must be used.
5.4. Combination of diffusion and Butler-Volmer kinetics
The combination of diffusion limitations and electrochemical kinetics can be used to obtain the following equation [58]:
Fig. 11. Comparison of rate and the square root of rotational velocity. The data come from copper dissolution in ferric sulfate aqueous media. (After Ref.
[59]).
h i
a FðEE Þ ac FðEEeq Þ
io exp a RT eq exp RT
i¼n h io ð50Þ
a FðEE Þ ac FðEEeq Þ
1 þ io i1 exp a RT eq i1 exp RT
la lc
where ila is limiting anodic current density and ilc is limiting cathodic current density. This equation fits experimental elec-
trochemical data well for cases involving oxidizing ions and bare metal surfaces under mass transport and/or electrochem-
ical rate control.
6. Surfactant
6.1. Introduction to surfactant
A widely used corrosion control method is to use organic surfactant inhibitors, many of which are surfactants with hydro-
philic and hydrophobic molecular sections [84–89]. One example of a surfactant molecule of homologous benzalkonium
(a)
(b)
H2O
Fe
(c) (d)
Fig. 12. Chemical formula of benzyl dimethyl hexadecyl ammonium chloride (C16, or C16Cl, or C16BzCl) (a) and the corresponding optimized molecular
geometry, (b) (After Ref. [89]). Dotted line region: functional group; dashed line region: hydrocarbon tail. Equilibrium adsorbed configurations on Fe (0 0 1)
surface of (c) dodecylamine (DDA) and (d) the protonated form of DDA (DDAH) in aqueous solution containing chloride ions (After Ref. [85]).
chlorides (BAC), hexadecyl trimethyl ammonium bromide (C16, C16Cl, or C16BzCl), is shown in Fig. 12(a) and (b) [89]. C16
has an N-based aromatic functional group which is hydrophilic, and a hydrophobic hydrocarbon tail with 16 linear CH2 and
CH3 sections. The hydrophilic group strongly prefers interaction with polar entities such as water or other ions, whereas the
hydrophobic section strongly prefers interaction with other hydrophobic entities such as hydrocarbons. This dual nature of
surfactants determines their interactions with surfaces and interfaces. Fig. 12(c) and (d) presents the equilibrium adsorption
on Fe (0 0 1) surface of dodecylamine (DDA) and the protonated form of DDA (DDAH) in aqueous solution containing chloride
ions. The information of all the surfactant compounds discussed in this review is summarized in Table 1.
There are three main categories of corrosion inhibitor mechanisms. Inhibitors are generally classified as barrier, anodic, or
cathodic inhibitors, although they commonly act by more than one mechanism. Barrier inhibitors reduce the rate of corro-
sion by creating a barrier to corrosion reactant transport or by blocking reaction sites on the surface. Anodic inhibitors inhibit
the anodic reaction. Cathodic inhibitors decrease the cathodic reaction at a corroding surface. It is reported that it is feasible
to classify an inhibitor as anodic or cathodic if the open circuit potential in the presence of inhibitor shifts at least +85 mV or
85 mV, respectively, relative to the open circuit potential in the absence of inhibitor [90].
Surfactant inhibitors can act as either anodic or cathodic inhibitors. Surfactants adsorb on metal surfaces and form a phys-
ical barrier that reduces corrosion. Consequently, surfactants are also often considered to be barrier type inhibitors. Surfac-
tant inhibitors often contain either a nitrogen or sulfur atom as part of the functional group of the molecule. Very commonly,
the nitrogen atom is part of a hydrocarbon ring, and its unbounded pair of electrons commonly forms a bond with metal
substrates. Similarly, the p electrons in resonating hydrocarbon ring structures interact with metal substrates to form a bond
with the substrate. This type of bonding often leads to bonds that are classified as chemical bonds that lead to the charac-
terization of associated adsorption as ‘‘chemisorption”. This type of adsorption makes it difficult for these molecules to be
displaced by reactants or to allow exposure of the metal surface to the environment at the adsorption site. However, it
should be noted that most metals used in corrosive environments have metal oxide surfaces that interact with surfactants
differently than metals.
Surfactant molecules must adsorb in order to be effective as corrosion inhibitors. Adsorption depends on the composition
of the solution, the concentration of the adsorbate, the interaction of the adsorbate with the surface, the properties of the
surface and the adsorbate, and the electrochemical potential of the surface. The types of organic molecule adsorption are
typically chemisorption, physisorption, electrostatic adsorption, and p-bond orbital interactions that have both physisorp-
tion and chemisorption properties. Inhibitors are further classified into proton acceptors, electron acceptors, and mixed
molecules [90]. Surfactants are important inhibitors that can act through several mechanisms due to their unique properties.
6.2. Hydrophilicity and hydrophobicity of surfactant
The hydrophilic functional group of surfactant molecules strongly prefers interaction with polar entities such as water,
metals, and other ions. Generally, surfactants adsorb on the metal surface, block the active sites exposed to corrosive media,
and thereby reduce corrosion attack [25,28,55,89,91]. It is believed that the structure of heterocyclic surfactant molecules
plays a dominant role in the corrosion inhibition. The presence and structure of specific atoms, such as N and O, in these
molecules strongly influences the adsorption mechanism and corrosion inhibition efficiency [55,91–94].
The hydrophobic portion, which is nonpolar, strongly prefers interaction with hydrophobic entities such as hydrocarbon
phase [89,91,95]. Therefore, surfactant molecules are prone to adsorb at and cover the surfaces/interfaces, such as air-liquid
surface and liquid-solid interface, to escape from polar solvent such as water by associating and packing hydrocarbon chains
together. The surfactant concentration at which a monolayer of surfactant molecules adsorbs on and covers metal surface is
termed the surface aggregation concentration (sac or satc) which is critical to corrosion inhibition [87–89,91,94,96,97]. As
surfactant concentration increases, bilayers/multilayers are likely to form on metal surface. Surfactant molecules can also
form aggregates in aqueous phase at solubility saturation in a way that they usually orient their hydrophobic tails toward
those of neighboring surfactant molecules and their hydrophilic head groups toward water or hydrophilic surfaces. The sur-
factant concentration at which surfactant molecules start to form aggregates such as micelles in solution is termed the crit-
ical micelle concentration (cmc) [87–89,91,94,96,97]. It has been shown that the sac is usually much lower than the cmc and
that high efficiency of corrosion inhibition is usually achieved at the sac provided that the surfactant is a good corrosion inhi-
bitor [89,91,94].
Metal and metal oxide surfaces are hydrophilic [98]. Consequently, the functional group in surfactant molecules is
attracted to surfaces of metals and metal oxides. The attraction to surfaces is strengthened by the hydrophobic portion of
the molecule that is attracted to other hydrophobic portions of adjacent surfactant molecules on surfaces. Thus, there is a
driving force for surfactant adsorption on metal and metal oxide surfaces that orients the surfactant with the hydrophilic
group at the solid interface and the hydrophobic hydrocarbon chain directed out into the solution, thereby creating a
hydrophobic surface. This driving force causes surfactant molecules to aggregate on surfaces. If sufficient surfactant is pre-
sent in solution a second layer or multiple layers of surfactant may adsorb creating a variety of adsorbed structures. Surfac-
tants behave in a similar way in solution with aggregate structures such as micelles forming at moderate concentration
levels above the cmc [99].
A view of the effect of concentration on adsorption and aggregation of surfactant molecules is shown in Fig. 13
[88,89,100,101]. When the surfactant concentration is low, there is only slight adsorption of surfactants on substrate. At
intermediate concentration, surfactants start to aggregate on the substrate but are loosely distributed. At the concentration
Table 1
The name, structure, symbol, and chain length (n) of surfactants (symbols are used in the following text.)
Surfactant name Structure Symbol n

n-[2-[(2-Aminoethyl) amino] ethyl]-9- AAOA 8
octadecenamide
n-Benzalkonium chloride (BAC) Cn, CnCl, or 12, 14,

CnBzCl 16
CnH(2n+1)-COO(CH2CH2O)12CH3 CnCOOE12 9
Polyoxythylene alkyl ether CnEn0 12, 16
n-Alkyltrimethyl ammonium surfactant CnTAX 10, 12,

(X = Br, Cl) 14, 16
Potassium alkanoate CnH(2n+1)KO2 9, 11
Dodecylpyridinium bromide DDPB 12

Br -
Octylglucoside OG 8
Sodium dodecyl sulfate SDS 12
trans-Cinnamaldehyde TCA –
Sodium bis-(2-ethylhexyl)- AOT –

sulfosuccinate
Cetylpyridinium chloride CPC 14
(n-1)
Primary alcohol ethoxylate CnOEn0 11
Table 1 (continued)
Surfactant name Structure Symbol n

Polyoxyethylene nonyl phenyl ether NPEn0 9
Polyethylene glycol esters – –
Linoleate – –
Dodecylamine DDA 11
N-based alkyl amines and derivatives CnNH2 6
CnN(CH3)2 4
CnN 6
(CH2CH2OH)2
CnNO 4
CnNH 10
(CH2)3NH2
air air air air
solution solution solution solution
substrate substrate substrate substrate
low concentration Medium (< SATC) at SATC (< cmc) Above cmc
specific adsorption/ isolated aggregate widespread aggregates on solution aggregation/

electrostatic formation/ cooperative substrate/initial multilayer multilayers on substrates
neutralization adsorption formation
Fig. 13. Comparison of surfactant aggregation in solution and on surfaces as a function of concentration. (After Refs. [89,90,101,102]). The SATC represents
the sac.
around the sac, a relatively uniform and porous film of adsorbed surfactants distributed on the substrate, where the sub-
strate is relatively well-protected from corrosion [89]. Further increase of surfactant concentration up to and beyond the
cmc, the growth of aggregates on substrate occurs mainly by bilayers/multilayers adsorption and hemi-micelles/micelles
aggregation, which slightly contributes to additional corrosion inhibition [89].
It is usually assumed that corrosion inhibition in the presence of low surfactant concentration (usually lower than micelle
formation concentration) can be represented by the number of active surface sites of substrate covered by surfactant adsorp-
tion [57,88,89,102,103]. More and more active surface area is covered by surfactants and protected against corrosion as sur-
factant concentration increases. Near the sac and the cmc, the metal surface is assumed to be covered by one monolayer and
multilayers/micelles of surfactants, respectively [57,88,89,102,103]. As it is previously mentioned that surfactants form
micelles at solubility saturation in aqueous phase, the surfactant may form reversed micelles in oil phase at a certain con-
centration which is termed the oil cmc (Co). The cmc in aqueous phase is termed the aqueous cmc (Cw). The overall average
concentration at which the micelle starts to form in the entire oil-water environment is termed the apparent cmc (Cap). The
surface coverage h is given by
iocorr icorr
h¼ ð51Þ
iocorr
where iocorr and icorr are the corrosion current density without and with surfactant inhibitors in solution, respectively.
6.3. Adsorption mechanism
The adsorption mechanism of surfactant is usually determined by the adsorption free energy DGoad , which is correlated to
the adsorption constant using the equation below [89,91,104]:

1 DG o
K ad ¼ exp ad ð52Þ
C mw RT
where Kad is the equilibrium adsorption constant, which is usually calculated based on various adsorption isotherms which
will be discussed in the later sections, Cmw is molar concentration of water which is 55.5 M, R is the gas constant, and T is
absolute temperature. Generally, a negative value of DGoad demonstrates that the adsorption of surfactant on the steel surface
is a spontaneous process and shows a strong interaction between surfactant molecules and steel surface [105,106]. If DGoad is
more positive than 20 kJ mol1, the interaction between surfactant and metal is often classified as physisorption due to
electrostatic interaction. When DGoad is more negative than 40 kJ mol1, the adsorption usually involves charge sharing
or transfer between the surfactant molecules and the metal surface to form coordination bonds, which is also classified
as chemisorption [106,107]. However, physisorption can sometimes be energetically favorable and significant whereas
chemisorption may sometimes have relatively weak binding energy due to various factors that influence adsorption
[108,109].
Because chemisorption involves a chemical reaction between the adsorbate and the surface to form a specific bond, it is
generally limited to a monolayer. In contrast, physisorption does not have a specific bonding requirement and can involve
multiple layers under some conditions. The high bonding energy associated with chemisorption requires elevated temper-
ature exposure to desorb chemisorbed molecules. Physisorbed molecules are readily desorbed at moderate temperatures or
by lowering related vapor pressures or concentrations. A summary of the properties of chemisorption and physisorption is
presented in Table 2 [82,106,107,110].
Surfactant adsorption is related to surface tension, wettability, and other phenomena, and its influence has been studied
in great detail [111–119].
6.4. Surface tension
Surface tension occurs when cohesive energy exists between molecules. Hydrogen bonds and van der Waals interactions
are present between water molecules in aqueous media. However, at the air-water interface, hydrogen bonds are not com-
plete above the interface. Similarly, van der Waals interactions are weakened at the interface because there are no interact-
ing molecules in the air phase. Therefore, molecules present at the air-water interface have fewer bonding opportunities and
greater available energy than those in the bulk phase. This excess energy is the basis for surface tension.
In aqueous media the hydrophobic portion of surfactant molecules has a tendency to move to available interfaces to
escape the undesirable polar solvent. Adsorbed surfactant molecules at the air-water interface decrease surface tension.
Thus, surfactant molecules are more active at the air-water interface. Correspondingly, the term ‘‘surfactant” is derived from
‘‘surface active agent” that lowers surface tension. Compounds that have higher cohesive energy at the air-water interface
than that between water molecules are expected to increase surface tension.
Table 2
Comparison of physisorption and chemisorption [82,106,107,110].
Physisorption Chemisorption
Adsorption enthalpy: <25 kJ/mole (negative) Adsorption enthalpy: >40 kJ/mole (negative)
Thickness: variable up to multilayers Thickness: variable up to monolayer
Reversibility: moderate temp./low vapor press. Reversibility: elevated temp.
Adsorption site: nonspecific Adsorption site: specific
Surface tension is related to the adsorption density as described mathematically in the Gibbs’ adsorption equation
[120,121]:

1 d!
Ki ¼ ð53Þ
RT d ln ai T;P
where Ki is surface excess or adsorption density of surfactant ‘i’ in number of molecules per unit area or in mol/m2, ! is sur-
face tension in force per unit length or in mN/m, and ai is activity of surfactant ‘i’. The Gibbs’ adsorption equation shows that
the adsorption density increases as the slope of surface tension versus the logarithm of concentration becomes steeper.
Adsorption density, when multiplied by the effective thickness of the surfactant layer to obtain volumetric concentration
units, indicates the concentration of surfactant (solute) at the surface is higher than in the bulk. In the bulk solution surfac-
tant is dispersed and more dilute than at the interface where it is aggregated and packed more densely. Thus, adsorption
density is effectively the local concentration of surfactant at the adsorption interface.
The surface tension of solution decreases gradually as surfactant concentration increases until it reaches a minimum pla-
teau value. As the interface becomes more packed, additional surfactant molecules tend to aggregate in solution as well as at
solid-liquid interfaces. Additional surfactant above the level needed for minimum surface tension is utilized to form aggre-
gate structures like micelles. The surfactant concentration needed to produce micelles is known as the cmc. The cmc gener-
ally represents the maximum concentration of surfactant monomers that can be dissolved in solution. The cmc is reached as
the surface tension reaches the minimum plateau level. Because of this aggregation phenomenon occurring at the cmc, the
Gibbs’ equation is not valid above the cmc. All surfactant in solution that is in excess of the cmc forms aggregate structures
such as micelles and acts as a separate phase.
Fig. 14 shows surface tension data of benzalkonium chlorides, including dodecyl trimethyl ammonium bromide C12 (or
C12Cl or C12BzCl), tetradecyl trimethyl ammonium bromide C14 (or C14Cl or C14BzCl), and hexadecyl trimethyl ammonium
bromide C16 (or C16Cl or C16BzCl), which can be used to determine the cmc [94]. As expressed in the Gibbs adsorption equa-
tion, the surface excess increases as water molecules are replaced by surfactant molecules at the surface provided that the
surface tension versus log10(concentration) increases the magnitude of its slope. Usually the change in slope is large at low
concentrations, and then remains nearly constant as the cmc is approached.
The surface tension of water at the air-water interface at room temperature is 72 mN/m [122]. Appropriately selected
hydrocarbon surfactants are usually able to decrease surface tension to around 35 mN/m. Surfactants with fluorocarbon
chain groups are capable of decreasing the surface tension to 25 mN/m [123]. Addition of salt can further reduce surface ten-
sion due to decreased repulsion between ionic head groups of surfactant molecules that enhances molecular packing [124].
6.5. Enhancement of wettability
The contact angle of a liquid droplet on a solid surface is the physical manifestation of the balance of three interfacial
tension values as presented in Young’s equation [125,126]:
!SO !SW
cos H ¼ ð54Þ
!WO
Fig. 14. Plots of surface tension of aqueous solution as a function of surfactant concentration at 40 °C: A – C12BzCl in 0.171 M NaCl solution; B – C12BzCl in
0.856 M NaCl solution; C – mixed C12BzCl, C14BzCl, and C16BzCl at ratio of 0.15/0.70/0.15 in 0.171 M NaCl solution [94]. The arrows pointing to x axis
indicate the cmc values. The cmc is usually defined as the concentration where the decrease of surface tension stops or switches to a very low slope [94].
The plots indicate what is likely occurring with respect to surfactant adsorption and aggregation as a function of surfactant concentration and surface
tension [94].
Fig. 15. Illustration of contact angle of water in oil on a solid substrate.
where !SO is the interfacial tension of solid-oil interface, !WO is the interfacial tension of water-oil interface, !SW is the inter-
facial tension at the solid-water interface, and H is the contact angle of the liquid droplet on the solid surface. A represen-
tation of these interfacial tensions is presented in Fig. 15.
Young’s equation indicates that a decrease in interfacial tension at the solid-water interface with the addition of surfac-
tant results in a decrease in the contact angle. A decrease in the contact angle is an indication of increased wettability. The
ability of a liquid to spread or wet a surface is described mathematically using the spreading coefficient [72]:
D! ¼ !SO ð!SW þ !WO Þ ð55Þ

Wettability is controlled primarily by the hydrophilicity of the surface. Surfaces that are very hydrophilic have very low
contact angles. The contact angle is a function of the surface tension as indicated in Young’s equation. Surface tension is a
thermodynamic property. Consequently, wettability is a thermodynamic result of the interactions of the associated phases.
Wettability can be controlled by controlling the surface tension values through the use of surfactants as well as through
selection of liquids and the solid interface.
Hydrophilic surfaces repel organic phases such as oils, and hydrophobic surfaces repel water. Consequently, when oil and
water are mixed, the water in the mixture will attract to hydrophilic surfaces such as steel, precipitated salts, and corrosion
product layers even though oil phases may be present. However, as surfactant adsorbs on a surface it can modify the wet-
tability. In some cases surfactants can change the character of a surface from one that is hydrophobic to one that is hydro-
philic or from a hydrophilic surface to a hydrophobic surface.
Wettability can be affected by the surface roughness. Previous research shows wettability is affected by roughness
[125,126]:
!S !SL
cos H0 ¼ f ¼ f cos H ð56Þ
!L
where H0 is the contact angle of a rough solid surface, f is a roughness factor that is greater than unity for rough surfaces, !S
is surface tension of solid, !L is surface tension of liquid, and !SL is interfacial tension between solid and liquid. It is well
known that surface heterogeneity (contamination due to the presence of attached fine particles), thin films, deformation,
etc. can affect the contact angle [126]. Surfaces can be made exceptionally hydrophobic or ‘‘super hydrophobic” by creating
microscale roughness on a normally hydrophobic surface.
6.6. Other properties influenced by surfactants
Surfactants and their natural tendency to aggregate in solutions and at interfaces influence other solution properties. The
osmotic pressure of a solution normally is influenced by surfactant concentration, but only significantly until it reaches the
cmc. Properties such as detergency are also influenced by the cmc in a dramatic way as shown in Fig. 16 [110]. Other prop-
erties such as solution conductivity are similarly influenced substantially by the aggregation of surfactant into micelles as
indicated in Fig. 16.
6.7. Langmuir-Blodgett films
Surface pressure, which is related to surface tension, and molecular surface area for monolayer films at the air-liquid
interface are commonly measured and manipulated using a Langmuir-Blodgett (LB) balance [127]. The LB technique can
be used to reveal interesting behavior of adsorbed surfactant molecules. The surface pressure of Langmuir-Blodgett Films,
PLB , is defined as [127,128]:
PLB ¼ !o ! ð57Þ
where !o is surface tension without an adsorbed layer, and ! is the surface tension with an adsorbed layer. The LB balance
can be used to measure surface pressure as the surface area of the adsorbed film is changed by surface compression using a
movable barrier. Thus, the LB method can be used to modify the area occupied per surfactant molecule using a movable com-
pression barrier.
Fig. 16. Comparison of relative property values and surfactant concentration for surface tension, detergency, osmotic pressure, and molar conductivity for a
typical ionic surfactant. (After Ref. [110]).
When the surface pressure in an LB film device is low and the surface area per molecule is high, surfactant molecules are
well dispersed and often are not oriented vertically at the interface. This well dispersed molecular state is often related to the
gas phase. Multiplying the surface pressure by the surface area leads to the ideal gas equation analog for a 2-dimensional
surface [82]:
PLB ALB ¼ nLB RT ð58Þ
where ALB is surface area of LB film and nLB is the number of moles of adsorbed surfactants on LB film. As the LB surface area
decreases surfactant molecules become more close-packed, and the surface pressure increases. This often leads to a molec-
ular state that is analogous to a liquid.
Further decrease in surface area forces tight molecular packing and the surface pressure reaches a maximum value. This
tightly packed condition is analogous to a solid phase because the molecules are packed and can have two-dimensional,
crystal-like structures. The Langmuir film balance can be used to produce oriented multilayer films by transferring the films
to substrates that are moved slowly through the air-water interface [70]. Thus, the Langmuir film balance can be used to
transfer surfactant films of varying packing density as monolayers or multilayers to any solid substrate. Consequently, prop-
erties such as hydrophobicity and diffusion barrier performance can be evaluated using these films as a function of molecular
packing and/or the number of multilayers on a substrate.
6.8. Krafft point
Surfactant molecules do not dissolve appreciably in aqueous solutions below a certain temperature, known as the Krafft
point [129]. Above the Krafft point or temperature, surfactants can form micelles, which greatly increases overall surfactant
solubility. A diagram illustrating the relationship between the Krafft point and monomer and micelle surfactant forms is pre-
sented in Fig. 17. Thus, the Krafft point provides an important boundary condition for surfactant utilization.
6.9. Surfactant states
Surfactants can organize into different phases or states. The phase or state of surfactant is generally closely related to con-
centration. If surfactant concentration increases above the cmc, monomeric surfactants form aggregate structures such as
micelles. Properties of micelles, micellization process, and associated modeling will be discussed in detail in subsequent sec-
tions. The quantitative relationships amongst the total surfactant concentration, monomer concentration, micelle concentra-
tion and the associated cmc are presented in Fig. 18. In addition to micelles, surfactants can form gels at the gel temperature
[125,130]. Surfactants can also form liquid crystals. The gel state is the transition between the liquid crystal phase and the
solid state. Surfactant in the gel state can change orientation and rotate. In contrast, surfactant dissolved in solution can have
full freedom unless it is part of aggregate structures.
Another interesting phase related phenomena occurs at elevated temperature for many surfactants. At elevated temper-
atures some surfactants have decreased solubility. This decrease is due to reduced hydrogen bonding resulting from higher
energy and conformational changes in surfactant structure that reduces bonding with water. Consequently, some surfactant
solutions become cloudy at elevated temperatures due to the reduced solubility above the ‘‘cloud point” due to secondary
phase formation. The cloud point effect is generally associated with nonionic surfactants.
Fig. 17. Comparison of total surfactant concentration and temperature and its relationship with surfactant states. (After Ref. [130]). Arrows next to the axes
point towards the increasing direction.
Fig. 18. Comparison of phase (micelle or monomer) concentration and total surfactant concentration for a hypothetical surfactant. Arrows next to the axes
point towards the increasing direction.
6.10. Micelles
Surfactant molecules are compelled to aggregate to form micelle-like structures above the cmc. A micelle generally con-
sists of tens to hundreds of surfactant molecules [109,131].
The number of surfactant monomers in a micelle is known as the aggregation number [109,131]. The molecular weight of
a micelle can be obtained simply by multiplying the aggregation number by the molecular weight of the surfactant
monomer.
The aggregation number of surfactant is affected by ionic strength, head group properties, hydrophobic chain dimensions,
and temperature. If there is a change in ionic strength of solution, the change can affect the aggregation number. Increasing
ionic strength weakens the repulsion between ionic head groups and increases the repulsion of the hydrophobic tail from the
aqueous medium, thereby enhancing the tendency for aggregation. The head groups of surfactant molecules determine the
outer size of a micelle. Small head groups facilitate aggregation. Temperature also influences aggregation [132]. Increasing
temperature can cause surfactant to dehydrate the hydrophilic group of surfactant, which can result in increased hydropho-
bicity. However, increased temperature also decreases the tendency to adsorb at interfaces.
Micelles are constantly interacting and exchanging with individual surfactant molecules. Individual monomer surfactant
molecules move in and out of micelles at the microsecond time scale [133]. In contrast, micelles form and dissociate at a time
scale of milliseconds. However, these time scales change very significantly as the concentration of surfactant and the size and
characteristics of the surfactant change.
Micelle aggregates form in shapes that include plate-like micelles, hemimicelles, spherical micelles, rod-like or cylindrical
micelles, and vesicles, etc. [109,134]. The governing factor of the shape of a micelle is the packing of the hydrocarbon chain of
the surfactant. Israelachvili shows a packing parameter (shape factor) will determine the tendency to form spherical micelles
or nonspherical micelles [109]. The packing parameter P s can be expressed as [109]:
vs
Ps ¼ ð59Þ
ao ls
where v s is a hydrocarbon chain volume, ao is an optimal surface area of surfactant headgroup, and ls is a critical chain
length. The preferred structure of micelles is summarized in Table 3 according to packing parameter [109]. A spherical
micelle is shown in Fig. 19 along with the dimensions used for the packing factor [109].
The aggregation number of a spherical micelle, N sph , and the aggregation number per unit length of a cylindrical portion of
Ncyl
the aggregate, ls
, are calculated from the aggregate radii r sph and rcyl [109,135]
4 p r 3sph
Nsph ¼ ð60Þ
3 vs
Ncyl prcyl
2
¼ ð61Þ
ls vs
rsph and r cyl are given, respectively, by [109,135]
3v s
rsph ¼ ð62Þ
ao
2v s
rcyl ¼ ð63Þ
ao
The surface charge density of a spherical micelle and of a cylindrical portion are given by [135,136]
2
q rsph
rsph ¼ ð64Þ
ao rsph þ r h
q r cyl
rcyl ¼ ð65Þ
ao r cyl þ r h
where q is the charge of the surfactant’s headgroup, and rh is the distance from the location of the charge of headgroup to the
surface of the micellar core. r h is specific to particular headgroup.
Tanford demonstrated that the critical chain length ls and hydrocarbon chain volume can be approximated by the follow-
ing equations [134]:
ls 6 lmax ﬃ ð0:154 þ 0:1265Li Þ nm ð66Þ
and
v s ﬃ ð27:4 þ 26:9Li Þ 103 nm3 ð67Þ
Table 3
Micelle structure as a function of packing parameter Ps [109].
Value of packing parameter Structure

Ps 1/3 Spherical micelle
1/3 < Ps < 1/2 Ellipsoidal micelle
Ps 1/2 Cylindrical micelle
1/2 < Ps < 1 Various interconnected structures
Ps 1 Vesicles and extended bilayers
Ps > 1 ‘‘Inverted” structures
Fig. 19. Spherical micelle in equilibrium with an individual surfactant monomer with associated parameters (After Ref. [109]).
where lmax is the maximum length of surfactant molecule and Li is the number of carbon atoms in the hydrocarbon chain of
surfactant ‘i’. Thus, the micelle packing parameter for DDPC (dodecyl pyridinium chloride) with an aggregation number of 80
is 0.37 (>0.33). A value greater than 0.33 indicates the DDPC micelle is likely to be nonspherical.
Micelles are generally considered to be a separate liquid phase. Micelles behave as a separate phase from a thermody-
namic perspective. Correspondingly, some compounds dissolve in micelle liquids [137,138]. Hydrophobic entities dissolve
in the hydrophobic interior of micelles in aqueous solutions. In organic media where micelles are inverted with polar inte-
riors, aqueous or polar entities are dissolved in the micelle interior.
6.11. Microemulsions
Microemulsions are similar to solutions with micelles. They consist effectively of swollen micelles that contain a 5–
100 nm droplet of liquid inside. They can exist in an oil continuous phase with nanosize water droplets inside of a surfactant
shell, or they can exist in a water continuous phase with oil droplets inside. Microemulsions, which have a wide variety of
uses, consist of alcohol with an ionic surfactant as well as water and oil. Microemulsions can also be synthesized using water,
oil, and a single surfactant such as sodium bis-(2-ethylhexyl)-sulfosuccinate, which is also known as AOT [139]. Microemul-
sions are thermodynamically stable forms of emulsions [140].
6.12. Surfactant partitioning
Surfactant molecules form different phases in water and oil and their mixtures. A general phase diagram is shown in
Fig. 20 [110]. This figure illustrates there are often several types of surfactant mediated structures that form in water-oil-
surfactant mixtures.
In a solution containing oil and water surfactant molecules can be found to be in equilibrium in several forms under cer-
tain conditions. Surfactant can partition to the oil and aqueous phases, it can adsorb on solid surfaces, it can form dimers and
micelles, it can combine with metals or hydrogen ions to form metal salts and hydrogenated compounds, and it can form
other compounds. It is anticipated that solving all of the equilibrium processes simultaneously can be used to determine
Fig. 20. Phase diagram for surfactant, water, and oil illustrating different phase regions for a typical alkane oil and alkyl surfactant. (After Ref. [110]).
the surfactant concentration for effective adsorption, which can then be used in combination with methods for corrosion
inhibition to predict the associated inhibition efficiency.
Associated modeling of micellization and partitioning are introduced in subsequent sections.
6.13. Common surfactant inhibitors
The most widely used cationic surfactants are quaternary ammonium salts and amines (when protonated), in which the
cation acts as a surface active specie [89,141]. These surfactants usually perform well as inhibitors in low pH solutions
because these surfactants should be in the protonated state and low pH is helpful to ensure protonation. It is reported that
certain pure and mixed long-chain quaternary ammonium bromides were used as efficient corrosion inhibitors for steel
materials [89,91,94,142]. Other typical surfactants commonly used in oilfields include alkyl phenol ethoxylate such as poly-
oxyethylene nonyl phenyl ether (NPE) [142–145], primary alcohol ethoxylate [146], polyethylene glycol esters [144], and
polyoxythylene alkyl ether [144].
6.14. Surfactant mixtures
It has been a practical observation that pure surfactant inhibitor is usually either not available at low cost or not effective
enough for corrosion inhibition and a proper mixture containing additional surfactants, intensifiers, solvents, and co-
solvents is desired [89,91,94,141,146,147]. In practical applications surfactant mixtures have received wide attention
because of their superior physicochemical properties and capabilities in efficient adsorption, solubilization, dispersion, sus-
pension, and transportation [89,148,149]. Solutions containing mixed surfactants can often be conveniently tuned to achieve
desired properties by adjusting the mixture composition. More surface-active and expensive surfactants are usually mixed
with less surface-active and less expensive surfactants to reduce cost [89,91].
It is believed that there is a synergistic effect of mixed surfactants on corrosion inhibition of metals [89,91,106,150],
which results in an improved performance of mixed surfactants relative to individual surfactants. The synergistic inhibition
has been shown to be an effective method of improving the inhibition efficiency, decreasing the amount of dosage, and diver-
sifying the application of surfactants [89,91,106,150]. In addition, a co-operative effect in corrosion inhibition of metals
occurs upon introduction of halide ions to corrosive media which contains surfactant inhibitors. However, the addition of
halide ions may either stimulate or inhibit corrosion of metal, depending on concentration. It has been reported that the inhi-
bitive effects of halides follow the order of I Br > Cl. The strong synergistic effect of iodide ion can be explained by the
chemisorption with metal surface because of its larger size and polarizability [106,151–153].
A synergistic parameter Ssn , is introduced to describe the combined inhibition behavior of amines and halide ions [150].
The synergistic parameter Ssn is determined as follows for the interaction of inhibitors 1 and 2 [150]
1 hcalc
1;2
Ssn ¼ ð68Þ
1 hmeas
1;2
where hmeas
1;2 is the experimentally measured surface coverage of mixed surfactant inhibitors 1 and 2. hcalc
1;2 is the calculated
surface coverage assuming no interaction between the inhibitors and is given by
hcalc
1;2 ¼ h1 þ h2 h1 h2 ð69Þ
where h1 and h2 are surface coverages of surfactant inhibitors 1 and 2, respectively. It is generally agreed that if Ssn
approaches 1 no interaction between the two inhibitors exists, if Ssn > 1 a synergistic effect applies, and that if Ssn < 1 an
antagonistic interaction predominates [150–155].
One example of corrosion inhibition utilization of surfactant mixture is shown in Fig. 21. According to the authors [89,94],
C16 has good corrosion inhibition for X65 steel but poor solubility in aqueous phase. In contrast, C12 is very easy to dissolve
in aqueous phase but can only provide good corrosion inhibition at much higher concentration (
two orders of higher) and
pure C12 is much more expensive than C16. The overall performance of C14 lies in between that of C12 and C16. In addition,
the cost to isolate individual surfactant from industrially produced surfactants which are usually mixtures of homologous
surfactants is another consideration. Therefore, it is wise to mix the three of them to achieve good corrosion inhibition at
the minimized economic investment. It is interesting to find that the corrosion current density is low at the sac, and addi-
tional increase in surfactant concentration above the sac does not contribute much to extra corrosion inhibition.
Appropriately mixed surfactant systems have also been shown to improve performance of some desirable interfacial
properties. In a study by Shiao [156], the effects of chain length matching of mixed surfactants on melting points, evapora-
tion retardation, micellar stability, foaming, lubrication, enhanced oil recovery, corrosion, and microemulsion formation
were discussed in detail. The results from this study showed that when an ionic surfactant is mixed with a nonionic surfac-
tant in a one to one ratio, the molecular packing is enhanced. Correspondingly, optimum properties such as corrosion resis-
tance, enhanced oil recovery, melting point, evaporation retardation, foam formation, and surface viscosity were affected by
the chain length compatibility. The rationale for the improvement in properties is that matching chain lengths allows for
more stable interactions between surfactant molecules in adsorbed layers [139,156–158].
Fig. 21. Plot of corrosion current density versus the total concentration of mixed BAC surfactants (C12/C14/C16 = 1:1:1) divided by the corresponding mixed
aqueous cmc Cw in 0.599 M NaCl aqueous solution at 40 °C and pH 5. (After Ref. [89]).
Certain additional studies on mixed surfactant inhibitor can be found elsewhere [159–161], such as the report using cin-
namaldehyde, benzalacetone, and chalcone with propargyl alcohol to protect steel from corrosion in 20% HCl at 90 °C [161];
however, the majority of the well-documented work [6,96,141,162–164] is focused on single compounds as corrosion inhi-
bitors for steel materials, and the corrosion inhibition is usually not effective enough or too costly for industrial applications.
Therefore, the development of a comprehensive theory or model is needed to adequately predict steel corrosion inhibition
using various mixed surfactants in WOS environments.
7. Various determining factors of inhibitor efficiency
Upon addition to a WOS environment, surfactant inhibitors can be involved in many processes, including adsorption on
steel surfaces, water/oil partitioning, precipitation, surfactant interactions, and micellization, etc., which, in turn, affect the
availability of monomeric surfactants in aqueous phase, which directly influences corrosion inhibition efficiency
[89,91,94,95,104,165]. Other factors, such as environmental temperature and fluid flow rate in pipeline, also need to be con-
sidered [95,104]. Therefore, it is not a straightforward task to evaluate the effect of available monomeric surfactants on metal
corrosion.
Many corrosion inhibition modeling approaches are available in the literature. These approaches vary from the use of
simple inhibitor factors and inhibition efficiency to the application of complicated molecular modeling techniques to
describe inhibitor interactions with the steel surface and iron carbonate scale [24,29,84–86,89,91,94,95,101,102,104,146–
155,166–177]. The most used approach is based on the assumption that corrosion protection is achieved by surface coverage,
which slows one or more electrochemical reactions, as mentioned previously [24,89,91,94,95,104]. The criteria to evaluate
the efficiency of pure or mixed surfactant inhibitors that are deployed in a WOS environment is summarized in six areas
[24,29,89,104,141,178–180]: (1) high corrosion inhibition efficiency g; (2) effective oil/water partitioning; (3) minimal ten-
dency of emulsion formation; (4) availability and stability of surfactant inhibitors; (5) materials compatibility; (6) feasible
monitoring and continuous injection of inhibitors. For comprehensive design, selection, utilization, and performance
improvement and prediction of surfactant inhibitors in an industrial WOS environment, the consideration of the above cri-
teria is highly recommended.
Various processes and phenomena which generally occur upon the injection of surfactant inhibitors to a WOS environ-
ment are shown in Fig. 22 [181]. Certain processes play a dominant role in the determination of corrosion inhibition, such
as adsorption on metal surface, surfactant aggregation, and oil/water partitioning, while other processes such as adsorption
at oil/water interface slightly affect corrosion inhibition. These processes and associated effects on corrosion inhibition will
be reviewed one-by-one as follows.
7.1. Adsorption at surface/interface
7.1.1. Adsorption isotherms

Surfactant adsorption is a prerequisite to surfactant-based corrosion inhibition. The relationship between adsorption and
corrosion inhibition has been studied by many researchers [9,13,24,73,74,88,89,91–97,99,136,170–174,181–208]. Surfactant
adsorption is often modeled using a variety of methods.
A common interpretation of the adsorption of ionic surfactant on oppositely charged substrates is depicted in Fig. 23 with
four adsorption regions (I, II, III, and IV) [101,209].
Fig. 22. Schematic representation of surfactant distribution: (a) cross-section of steel pipe containing water, oil, and limited amount of vapor of oil and
water; (b) schematic illustration of surfactant (cationic surfactant as an example) distribution and various processes in a WOS environment with dissolved
CO2 at the average surfactant concentration above the apparent cmc [181].
In this view, the adsorption trends are linear in Region I due to electrostatic attraction. The adsorption density follows the
Gouy-Chapman equation with a constant slope under constant ionic strength conditions. The ions’ adsorption density (neg-
ative ions as an example) is given by [209]
Fig. 23. Qualitative comparison of adsorption density and residual ionic surfactant concentration on logarithmic scales with associated surfactant
adsorption structures for an oppositely charged substrate. (After Ref. [101]).
r
K ¼ ð70Þ
zF
where r is surface charge density of solid surface/substrate and hardly change under constant ionic strength conditions.
Increased adsorption at higher surfactant concentration leads to Region II in which the adsorption density increases more
than the corresponding solution concentration due to aggregate structure formation at the surface. The adsorbed surfactant
can be in the form of hemi-micelles due to lateral interactions between hydrocarbon chains that can lead to patches of
packed molecules [209].
In Region III, electrostatic attraction is no longer operative due the electrically neutralized solid surface by the adsorbed
surfactant, and adsorption takes place due to lateral attraction alone with a reduced slope. Region IV is a saturated region
with not much additional adsorption as the residual solution concentration increases because of solution micelle formation.
The adsorption in this region is mainly through lateral hydrophobic interaction between the hydrocarbon chains. The local
bilayer areas may be in the form of cylindrical or rod-like micelles, in which surfactant molecules adsorb with a reversed
orientation (head groups facing the bulk solution) resulting in a decrease in the hydrophobicity of the particles in this region
[101].
Nonionic surfactants usually contain polar headgroups, which tend to form hydrogen bonds with the hydroxyl groups on
the solid surface/substrate. The adsorption of nonionic surfactants to most solids is weaker than that of ionic surfactants con-
sidering that hydrogen bonding is weaker than electrostatic interaction. Therefore, nonionic surfactants exhibit a sharp
increase in region III in the adsorption isotherm due to the absence of electrostatic interactions. In other regions, nonionic
surfactants and ionic surfactants exhibit similar characteristics in the adsorption isotherms [101].
Corrosion inhibition is directly determined by the effective adsorption of surfactant monolayers and bilayer/multilayers
on the steel substrate/surface due to the physical and chemical blockage of the surface active sites exposed to corrosive
media [89,91,94,95,102,181]. Physisorption is usually accomplished through van der Waals forces and electrostatic interac-
tions between polar or charged functional groups and charged/polar steel (or other metal substrate) surface [89,91,94,95].
The adsorbed surfactant chemically modifies steel in a way that the functional groups partially donate electrons to iron
and link the steel substrate by forming a partial chemical bond, leaving the hydrocarbon tails pointing outwards and forming
a densely packed hydrophobic barrier, which is believed to inhibit the diffusion of water, carbonate ions, halide ions (if there
is any), hydrogen ions, and oxygen, etc. to the surface [181]. Adsorption behavior can usually be evaluated using experimen-
tal methods [210–214], such as Fourier transform infrared spectroscopy (FTIR) and polarization modulation infrared reflec-
tion absorption spectroscopy (PM-IRRAS) [212–214], or using computational approaches such as density functional theory
(DFT) and classical molecular dynamics simulations [13,92,93,169–172,200,202,204,215–218].
The adsorption of a surfactant molecule at a steel surface (it is actually steel/solution interface) can be presented as a sub-
stitution adsorption process between the surfactant molecules in aqueous solution, (Saq), and the water molecules on the
metallic surface (H2Oad) [219,220]:
Saq þ 1 H2 Oad $ Sad þ 1 H2 Oaq ð71Þ
where 1 is an empirical fitting parameter, which is interpreted as the number of water molecules displaced by one surfactant
molecule.
Surfactant adsorption is dependent on adsorption energy, lateral interactions of surfactant molecules, and associated
entropy change [89,104,167]. As discussed previously, the corrosion inhibition efficiency is assumed to be equal to the effec-
tive surface coverage, and thus adsorption models have been constructed from the change of the corrosion rate or equivalent
parameters such as polarization resistance and charge transfer resistance from electrochemical measurements as a function
of surfactant concentration [89,91,94,95,221]. The most frequently used adsorption models are the Langmiur model, Temkin
model, Freundlich model, Frumkin model, Flory-Huggins model, Dhar-Flory-Huggins model, Bockris-Swinkels models,
Bockris-Devanathan-Muller model, Van der Waals-Stern model, and Stern Adsorption model [24,70,80,82,89,91,103,193,1
12,222–226], which are described below.
7.1.1.1. Langmuir isotherm. The Langmuir adsorption isotherm assumes that all surface adsorption sites are equivalent-
regardless of whether neighboring sites are occupied. Using the Langmuir model, the adsorbed adsorbate concentration is
expressed as:
h
K ad C w
m ¼ ð72Þ
1h
where h is surface coverage, K ad is the adsorption equilibrium constant, and C w m is the total concentration of monomeric sur-
factants in aqueous phase or available surface cites. Determination of whether or not the Langmuir model applies to a given
set of data can be made by plotting 1=h vs. 1=C w m and evaluating the linearity of the data. The equilibrium adsorption con-
stant, K ad , can be calculated from the slope. K ad can be used to calculate the free energy of adsorption as discussed previously.
Some of the limitations of the Langmuir model are the neglect of molecular interactions, inability to account for multilayer
coverage (along with all other general models) and the neglect of heterogeneous surface sites. However, despite its limita-
tions, it is generally very effective and well used.
The associated equation for a multicomponent system expressed as a fraction of sites occupied (hi ) is [70]:
K adi C w
hi ¼ P mi ð73Þ
1 þ i K adi C w
mi
where hi represents the surface coverage from surfactant ‘i’, K adi is equilibrium constant of surfactant ‘i’, and C w
mi represents
the monomeric concentration of surfactant ‘i’ in aqueous solution.
7.1.1.2. Temkin isotherm. The Temkin adsorption isotherm is an empirical adsorption model that considers nonuniform site
distribution. According to this model, the adsorbed adsorbate concentration is expressed as:
K ad C w
m ¼ expðnhÞ ð74Þ
where n is the molecular interaction constant (n < 0 indicates lateral attraction interactions between adsorbed surfactant
molecules; and n > 0 indicates lateral repulsion interactions between adsorbed surfactant molecules). n takes into the
account of adsorbent-adsorbate interactions. This model usually assumes that heat of adsorption of all molecules in the layer
decreases linearly rather than logarithmic with h when the extremely low and high values of concentrations are ignored
[227].
7.1.1.3. Freundlich isotherm. The Freundlich adsorption isotherm is derived by assuming a site distribution function that is
based on varying adsorption site free energy values. Using this model, the surface coverage can be expressed as:
1=1
K ad C w
m ¼h ð75Þ
where 1 represents the number of water molecules displaced by one surfactant molecule as previously described. It reduces
to the Langmuir equation for 1 = 1 at low concentrations and for 1 = 1 at high concentrations. The value of 1=1 is a measure
of adsorption intensity or surface heterogeneity, indicating more heterogeneous surface as its value approaches zero. A value
of 1=1 below unity implies chemisorption whereas 1=1 above one is an indication of cooperative adsorption [228].
7.1.1.4. Frumkin isotherm. The Frumkin isotherm can be represented by the equation:
h
K ad C w
m ¼ expðnhÞ ð76Þ
ð1 hÞ
Similar to Temkin isotherm, Frumkin isotherm also considers lateral interactions between adsorbed surfactant molecules
through n. Determination of whether or not the Frumkin model applies to a given set of data can be made by plotting
lnðh=ðC w
m ð1 hÞÞÞ vs. h and evaluating the linearity of the data [212].
7.1.1.5. Flory-Huggins isotherm. Flory-Huggins isotherm can express the feasibility and spontaneous nature of an adsorption
process and takes the following format:
h
K ad C w
m ¼ ð77Þ
1ð1 hÞ1
The linear form of Flory-Huggins model can be obtained by plotting lnðh=C w
m Þ vs. lnð1 hÞ [229].
7.1.1.6. Dhar-Flory-Huggins isotherm. Dhar-Flory-Huggins isotherm is similar to Flory-Huggins model except for one expo-
nential factor and has the following form:
h
K ad C w
m ¼ ð78Þ
ð1 hÞ1 expð1 1Þ
The linear form of Dhar-Flory-Huggins model can be checked by plotting lnðh=C w m Þ vs. lnð1 hÞ. It has been shown that it is
h
strictly incorrect to refer to an isotherm containing the configurational term 1ð1hÞ 1 as the ‘‘Flory-Huggins isotherm,” since
Flory-Huggins statistics leads to an isotherm having a different form for the configurational term, which is expð1 1Þ [230].
7.1.1.7. Bockris-Swinkels isotherm. Bockris-Swinkels isotherm was originally proposed by Bockris and Swinkels for the eval-
uation of adsorption of organic compounds on metal electrode and takes the following form [231]:
h ðh þ 1ð1 hÞÞð11Þ
K ad C w
m ¼ 1 ð79Þ
ð1 hÞ 11
Note that Bockris-Swinkels isotherm reduces to Langmuir isotherm when 1 ¼ 1.
There are other isotherms available, such as Van der Waals-Stern model [112], Stern Adsorption model [70], and Hill-de
Boer model [232], but they are not used as frequently as the above isotherms in the evaluation of surfactant adsorption on
metal electrode. Based on the experimental data, an appropriate adsorption isotherm can be selected for a particular surfac-
tant inhibitor of interest and the associated applications are provided elsewhere [89,91,103,212,229,231,233,234]. Please
note that all the isotherms are based on best-fit of experimental data, and they are only partially theoretically sound except
the Langmuir isotherm. The best-fit empirical parameters 1 and n usually cannot be extrapolated to other surfactants which
include homologous and nonhomologous surfactants.
7.1.1.8. Recent progress in adsorption isotherms. Recently a multi-interaction (MI) isotherm which describes monolayer
adsorption and lateral interactions between adsorbed surfactant molecules and the formation of surface aggregates based
on the combination of Langmuir isotherm and the aqueous cmc has been developed and described as follows [235]
w 1
Cw Cm
m
Cw Cw
Ke ¼ Kmax;1 Cw
þ Kmax;2 w 1 ð80Þ
K haf;1 þ Cwm
K haf;2 þ Cm
Cw
where Ke is the equilibrium amount of adsorption concentration (107 molecules/colony forming unit (CFU)); Kmax;1 and
Kmax;2 are the maximum adsorption concentrations for the two interactions; K haf;1 and K haf;2 are half saturation constants
for each interaction (unit less); Cw is the aqueous cmc. The first term on the right-hand side of the above equation is a Lang-
muir isotherm describing monolayer adsorption on the substrate surface, and the second term accounts for lateral interac-
tions between the adsorbed surfactants and formation of the surface aggregates. The multi-interaction isotherm adsorption
has been validated for linear polyoxyethylene (POE) alcohol surfactants of the form CxEy onto the surface of a Sphingomonas
sp. [235] and an example of the model application to C12E9 is given in Fig. 24. The fitting of an MI isotherm is excellent over
the entire concentration above and below the aqueous cmc while Langmuir isotherm fails to fit well. However, please note
that the MI isotherm has three best-fit parameters (1, K haf;1 , and K haf;2 ) and that these parameters have the same limitation as
those in the regular adsorption isotherms discussed above. Correspondingly, the extrapolation of the fitting parameters to
other surfactants usually leads to unreliable results.
As mentioned in the former section, the aqueous sac (represented using C) and cmc (represented using Cw ) are important
parameters characterizing corrosion inhibition efficiency of surfactants. Therefore, a new adsorption isotherm termed the
modified Langmuir adsorption (MLA) has been reported by incorporating the aqueous cmc into the regular Langmuir model
to evaluate surfactant adsorption and corrosion inhibition efficiency of surfactants under various solution conditions [87–
89,91,94,95,136,181]. The concentration range is usually confined between zero and the cmc for accurate evaluation. The
MLA is presented below
Fig. 24. Comparison of experimental and fitted adsorption isotherms for C12E9 onto the Sphingomonas sp. (After Ref. [235]).
1 Cw
¼ 1 þ K0 m ð81Þ
1h Cw
where K0 is equal to the adsorption constant Kad multiplied by Cw . Note homologous surfactants tend to achieve similar levels
of surface coverage at similar ratios of surfactant concentration to surfactant cmc, so the value of K0 for homologous surfac-
tants is relatively constant and can be used as a universal constant for such homologous surfactants [89,91,94,95]. Note that
m can increase up to the aqueous cmc C or above, but the deviation of the predicted data from experimental data will
the C w w
increase to some extent (within acceptable range). The deviation for the prediction of C w
m below the sac is higher, as shown in
following sections, which show that the sac is a transition point in characterizing the effectiveness of adsorbed surfactants
for corrosion inhibition. On the other hand, the corrosion inhibition is usually high enough at the sac, and therefore, the con-
tinuous increase of surfactant concentration up to the cmc or above does not contribute much to additional corrosion inhi-
bition and thus the deviation between MLA prediction and experimental results at or above the cmc should be small.
The essence of MLA is that the incorporation of cmc can successfully adjust for the effects of solution conditions and sur-
factant properties, such as salt concentration, solution temperature, hydrocarbon chain length, lateral surfactant interac-
tions, and counterion binding, on surfactant adsorption and thus on corrosion inhibition efficiency. It is interesting to
note that the regression parameter K0 in MLA for one surfactant can be transferred (extrapolated) to the corresponding
homologous surfactants and other surfactants with similar head groups (usually characterized by quantum descriptors)
[89,91,94,95]. It is therefore easy to understand that the best-fit parameter of K0 of any surfactant mixture can be used
for mixtures of similar surfactants [89,91,94,95].
The plots of MLA and some commonly used adsorption models based on the electrochemical measurements for a surfac-
tant mixture (C12/C14/C16 = 0.70/0.25/0.05 in 0.171 M NaCl aqueous media with CO2 saturation and pH = 4 at 40 °C) are
1
presented in Fig. 25, in which only MLA shows clearly the feature of the aqueous sac [89,91,94,95]. The MLA plot of 1h
w
vs. CCmw yields a slope of fit parameter K 0 = 13.74, and an intercept of 1 which is in the absence of surfactant inhibitors, as shown
in Fig. 25(d). There is one abrupt transition around the concentration of the aqueous sac, which indicates that when the sur-
factant concentration is below the aqueous sac, surface coverage increases rapidly with the increases in surfactant concen-
tration; above the sac, the increase in concentration does not contribute much to further surface coverage increase. The
K 0 = 13.74 can be extrapolated to other mixtures of BAC surfactants and provides predicted surface coverage and corrosion
inhibition, which is comparable to experimental data, as shown in Fig. 26 [89,91,94,95].
It has been shown that the amount of surfactant adsorbed at the water-oil interface, which is orders of magnitude less
than that of monomeric surfactants in the bulk phase provided that there is no turbulent flow of water-oil mixture, is unli-
kely to significantly impact the mass balance [236] and therefore, the amount of surfactants adsorbed on water-oil interface
and its effect on the availability of monomeric surfactants in the bulk solution for adsorption on steel surface and associated
corrosion inhibition can be neglected.
7.1.2. Adsorption thermodynamics

The standard free energy of surfactant adsorption for the adsorption-solution equilibrium is given by (see Eq. (71)) [91]
aSad CS
DGoad ¼ RT ln ¼ RT ln ad ð82Þ
aSaq C Saq
Fig. 25. The adsorption isotherms on X65 steel electrode of mixed BAC (C12/C14/C16 = 0.70/0.25/0.05) in 0.171 M NaCl aqueous media with CO2 saturation
and pH = 4 at 40 °C: (a) Langmuir adsorption; (b) Freundlich adsorption; (c) Temkin adsorption; (d) modified Langmuir adsorption (MLA). (After Refs.
[89,94]).
Fig. 26. The comparison of experimentally determined surface coverage and predicted surface coverage based on MLA and extrapolated parameter
K 0 = 13.74 on X65 steel electrode of mixed BAC (C12/C14/C16 = 1/1/1) in 0.599 M NaCl aqueous media with CO2 saturation and pH 5 at 40 °C. (After Refs.
[89,94]).
In this equation ‘‘a” is the activity and ‘‘C” represents the concentration of the specified form of the surfactant. Note the
activity coefficient for the aqueous and adsorbed forms is assumed to be equivalent, thereby cancelling its effect in the equa-
tion. Thus, the equation can be rearranged to:

DGo
C Saq exp ad ¼ C Sad ð83Þ
RT
The concentration of adsorbed surfactant is effectively the surface excess concentration per unit area, K, divided by the
thickness of the adsorbed surfactant layer, which is the same scale as the length of the surfactant molecule, lm . The concen-
tration of surfactant in the aqueous phase is effectively the bulk concentration of surfactant. Therefore, the equation for equi-
librium adsorption can be rewritten as [70]:

DGo
K ¼ C Saq lm exp ad ð84Þ
RT
Correspondingly, the surface excess at various surfactant concentration levels can be determined. Furthermore, the
change in free energy associated with adsorption can be calculated using the rearranged form of the equation:
DGoad ¼ RT lnðC Saq lm Þ þ RT ln K ð85Þ

Alternatively, the free energy of adsorption can be determined from the equilibrium adsorption constant [91]:
DGoad ¼ RT lnðC mw K ad Þ ð86Þ

Thus, the free energy of adsorption can be determined from equilibrium adsorption data. Temperature has a strong influ-
ence on surfactant adsorption. The relationship between free energy and temperature can be written in a traditional ther-
modynamic format:
DGoad ¼ DHoad T DSoad ð87Þ

Because surfactant adsorption is commonly an exothermic process, the enthalpy term in this equation is usually negative.
In order to have the required negative change in free energy for the reaction, the entropy term must be smaller than the
enthalpy term because the entropy for adsorption of surfactant is negative due to the ordering of the system associated with
adsorption of surfactant. Thus, if temperature increases, the entropy term, which has an overall positive sign, eventually
becomes more dominant than the negative enthalpy term, and desorption occurs. This same effect is manifest in kinetics.
Temperature also increases the rate of adsorption, but the corresponding increase in the desorption process becomes more
dominant in relation to the net equilibrium. Consequently, increasing the temperature generally decreases the ability of sur-
factant molecules to inhibit corrosion.
The enthalpy of adsorption is related to the change in equilibrium adsorption constant with respect to the change in tem-
perature [70]:

@lnK ad DHoad
¼ ð88Þ
@T h RT 2
Differentiation for constant coverage and comparison to enthalpy change leads to [70]:

@ ln C Saq @ ln K ad DHoad
¼ ¼ ð89Þ
@T h @T h RT 2
The change in the concentration or molality of adsorbate as a function of temperature at constant coverage is proportional
to the enthalpy change [70]:
!
@ ln C Saq DHoad
¼ ð90Þ
@ 1T R
h
Consequently, a plot of the logarithm of bulk surfactant concentration versus the change in inverse temperature has a
slope equal to the enthalpy change associated with adsorption divided by R.
The entropy of adsorption is influenced by the ordering that occurs when molecules adsorb in an ordered structure on a
surface. However, the adsorption of surfactant on a surface is accompanied by desorption of water molecules and adsorbed
ions that become disordered as they leave the surface.
One approach to determine the enthalpy of adsorption and the entropy of adsorption is to perform adsorption tests at
different temperatures then plot the free energy of adsorption versus temperature. Because free energy is related to enthalpy
and entropy, a plot of adsorption free energy versus absolute temperature provides a slope equal to the entropy of adsorption
and an intercept equal to the enthalpy.
7.1.3. Adsorption kinetics

The corrosion inhibition efficiency of adsorbed surfactants is affected by the packaging efficiency of surfactant molecules
and their competition with other species that promote corrosion (water molecules, halides, and organic acids, etc.). Reports
are available using mechanistic modeling [89,91,94,95,104] and molecular modeling perspectives [237–239] regarding the
packing efficiency, however, none of these studies assesses the kinetics aspect of surfactant adsorption which actually affects
the stability and availability of adsorbed surfactants in a way that if the adsorbed molecules are loosely packed with pores
the penetration of corrosive species may occur and promote corrosion [9,104,181,240–242]. Therefore, the comprehensive
evaluation of the transport of water, halides, carbonate ions, hydrogen and metal-complexes, etc. through the porous
adsorbed surfactants in the kinetics aspect should be performed. On the other hand, the surfactant concentration in aqueous
solution will decrease as a function of time due to their dynamic nature and the surfactant molecules may desorb. Under-
standing adsorption and desorption kinetics of surfactants is critical in the optimization of injection frequency of surfactants
to ensure effective corrosion inhibition [9,104,243,244]. The electrochemical corrosion inhibition measurement tests which
are performed in the laboratory are often performed to identify inhibited rates rather than kinetics, so often kinetics data is
missing.
The extent and rate of adsorption of many surfactant inhibitors are affected by the concentration of the inhibitor as well
as the concentration of competitive ions. As an example, the rate of sodium oleate adsorption on a fluorite surface is reduced
significantly by the presence of competitively adsorbing ions such as hydroxide, carbonate, and fluoride ions. Measurements
made using Fourier Transform Infrared internal reflection spectroscopy (FTIR/IRS), a continuous flow cell, and a fluorite inter-
nal reflection element showed that anions strongly influence the rate of oleate adsorption on fluorite [119,199]. These data
show that hydroxide has the most pronounced effect on adsorption kinetics with significant competitive adsorption in the
0.0001 M concentration regime, followed by carbonate, then fluoride [119,199].
Adsorbed surfactant molecules become more oriented as they pack tighter. In a comparison of LB monolayers and self-
assembled (SA) monolayers of stearate, Jang and Miller found that stearate molecules were oriented 9–16° from the surface
normal for LB monolayers and 21° for SA monolayers [245]. These conclusions are based on linear dichroism theory and the
associated use of polarized light infrared spectroscopy with different polarization angles to determine the molecular orien-
tation. Spectroscopic information was also used to determine the stearate molecules predominantly in the transition state. It
is also interesting to note that the orientation angle was nearly independent of adsorption density, suggesting that adsorp-
tion at low levels results in nearly vertically oriented molecules.
For a first order reaction described by Eq. (90), it can be shown that [119,246]:
dh
¼ kf C Saq ð1 hÞ ð91Þ
dt
where represents the surface coverage, kf represents the forward reaction rate constant, t represents reaction time, and C Saq
represents the bulk concentration of adsorbing species. Integration of Eq. (91) leads to [119,246]

1
ln ¼ kf C Saq t ð92Þ
1h
assuming no coverage at time zero. Thus, for constant solution concentration, plotting lnð1=ð1 hÞÞ vs. t should yield a slope
of kf C Saq and a zero intercept if the kinetics follow Eq. (92) assuming a first order reaction with respect to the adsorption
density at constant bulk concentration. Fig. 27 illustrates such a plot for the adsorption of 1 105 M oleate on fluorite in
which the maximum monolayer-level adsorption density was selected between the realistic monolayer coverage values
of 5.8 and 6.8 mol/m2 (based upon LB film results [199]) such that the best fit of the data was obtained as determined by
maximizing the R-squared values. As predicted by Eq. (92), a reasonably linear relationship exists between lnð1=ð1 hÞÞ
and time as illustrated in Fig. 27. Also, Fig. 27 shows the linear relationship is accompanied by an intercept near zero, though
it should be noted that for the regression analysis the line was forced through zero. The first order kinetics seems to be suit-
able at a concentration of 1 105 M oleate.
Chen and Frank have shown that adsorption kinetics follow a modified Langmuir based kinetic model that is expressed as
[70,247]:
dh kf kb
¼ C S ð1 hÞ h ð93Þ
dt C o aq Co
where kf represents adsorption rate constant, kb represents desorption rate constant, and C o is the concentration of adsorbed
surfactant at full coverage.
A kinetic adsorption model for an expanding air-water interface has been developed by Valkovska et al. [112]. This model
incorporates Fick’s second law of diffusion with traditional adsorption isotherm modeling to predict adsorption at an
expanding air-water interface. Valkovska et al. tested this model with alkyl trimethyl ammonium bromide surfactants in
0.1 m sodium bromide solution and found reasonable agreement between the model and the measured data. This model
is very complex and does not seem to fit data any better than simpler models. However, it does consider micelle breakdown,
and it may be useful for some applications with rapidly changing interfacial areas.
The process of competitive adsorption can significantly affect surfactant adsorption and desorption. A view of this process
is represented in Fig. 28 for an example of the competitive adsorption of cationic surfactant and ferric ions on chemical vapor
deposited tungsten (CVD) [248]. The competitive adsorption process is represented in Eqs. (94) and (95). These reactions
Fig. 27. Plot of lnð1=ð1 hÞÞ versus time for linoleate adsorption at a fluorite surface as measured from in-situ FTIR/IRS at 25 °C. The linoleate concentration
was 1 105 M. The maximum adsorption density was set at 6.2 lmole per m2. (After Ref. [119]).
Fig. 28. Illustration of competitive adsorption of cationic surfactant and ferric ions on chemical vapor deposited tungsten. (After Ref. [248]).
compete with each other for surface sites (W). The analysis follows methods applied in gas adsorption modeling [80]. The
reactions shown as Eqs. (94) and (95) can be written in terms of equilibrium constants K sur and K ion for cationic surfactant
and ferric ions, respectively. These constants represent the relative forward (adsorption) and backward (desorption) rate
constants. These constants are shown mathematically by Eqs. (96) and (97).
W þ Sion $ W Sion ð94Þ
W þ Ssur $ W Ssur ð95Þ
½W Sion
K ion ¼ ð96Þ
½W½Sion
½W Ssur
K sur ¼ ð97Þ
½W½Ssur
Sion represents Ferric ions or other competing ions. Ssur represents the cationic surfactant or other adsorbates. ‘‘[ ]” represents
concentration. ½W tol represents the total concentration of substrate surface sites. Solving Eqs. (96) and (97) with a surface
site balance leads to [248]:
K sur ½Ssur ½W tol
½W Ssur ¼ ð98Þ
1 þ K ion ½Sion þ K sur ½Ssur
This equation shows that the concentration of adsorbed surfactant depends on the concentration ratio of surfactant to
competitive ion as well as the equilibrium constants. When K sur is small and K ion is relatively large, the adsorbed surfactant
concentration is linearly related to the ratio of the surfactant concentration to the competitive ion concentration.
7.2. Surfactant aggregation and the aqueous cmc
As discussed in the former section, the incorporation of the aqueous cmc into MLA provides a substantial improvement in
the modeling of surfactant adsorption in that this method can describe surfactant adsorption on substrate surface and
account for lateral interactions among the adsorbed surfactants, formation of aggregates, and the environmental effects such
as salt concentration, solution temperature, etc. Therefore, the accurate evaluation of the aqueous cmc of pure and mixed
surfactants of interest is critical to the application of MLA. On the other hand, the aggregation process consumes most of
surfactants added to the aqueous phase above the aqueous cmc, which inevitably affects the availability of monomeric sur-
factants for adsorption on metal surface and associated corrosion inhibition [9,95,181].
Assuming the monomeric (ionic) surfactant mi (i = 1, 2, or 3. . .) is completely dissociated in aqueous solution containing
counterion mj (j = 1, 2, or 3. . .) but in the micelle form the surfactant is associated to some extent with counterions, the sur-
factant micellization can be described by the following process [9,136,249,250]
X X
Nð ai zi þ dj z j Þ
X zi
X z i j
N a i mi þN dj mj j $ MN ð99Þ
i j
where ai is the molar fraction of surfactant i in the micelle, MN, which has an aggregation number N, micelle composition ai,
and an ion binding coefficient dj [9,136,249–252]. For micelles of pure surfactant, ai = 1; for mixed micelles, 0 < ai < 1. zi and
zj are the valences of ionic surfactant i in dissociated form and ion j. For nonionic surfactant i, zi = 0 and dj = 0.
One of the challenges in the study of the aqueous cmc comes from the effects of specific ions and added salts on the aggre-
gation properties of surfactants. Different counterions usually have different effects on the aqueous cmc, micelle shape,
micelle growth, micelle size and distribution, mixed micelle composition (for mixed surfactants), and phase separation
[9,136,250,253–260]. It is reported that the counterion effect on the aggregation properties of cationic surfactants is usually
stronger than that of anionic surfactants [257,258]. In addition, the cmc depression due to the counterion effect usually fol-
lows the Hofmeister series: OH < F < Cl < Br < NO
3 < ClO3 < I < benzoate < salicylate

for cationic surfactants; and
Li+ < Na+ < K+ < Cs+ for anionic surfactants [136]. The specific counterion effects on micelle size and sphere-to-rod transition
is usually in the same order as shown previously for cmc [258]. The counterion binding mechanism, however, is not clear and
has been a controversial issue [261].
At low salt concentration, the coion effect on cmc, aggregation number, and sphere-to-rod transition is negligible
[136,262,263]. However, as salt concentration increases, the coion effect becomes increasingly noticeable [136,262,263]. Par-
ticularly at relatively high salt concentration, the coion effect on aggregation properties becomes dramatic, as discussed in
the text below.
In the approach proposed by Nagarajan and Ruckenstein [264] based on the work from Evans and Ninham [265], Nagara-
jan [264] successfully incorporates a parameter which is the distance between the surface of a hydrophobic/micellar core
(the micellar core is the micelle with hydrophobic chain and without headgroup) and the center of counterions. This param-
eter, according to Nagarajan [265], is dependent on headgroup size, hydrated counterion size, and the distance from the
counterion to the charge of ionic surfactant. However, Nagarajan does not provide more details about these dependencies.
The molecular thermodynamic theory (MTT) [254,266–268] provides a great step in progress toward the understanding and
modeling of counterion specificity on ionic surfactant micellization. In the theory, the counterion is assumed to bind to the
micelle surface in terms of fractional coverage between 0 and 1, and it affects the magnitude of various free energy contri-
butions to the micellization process. The predicted cmc as well as some other properties such as aggregation number and
mixed micelle composition are in relatively good agreement with experiment. However, the theory does not clarify the
specificity of headgroup-counterion pair interactions. More recently, Moreira and Firoozabadi (MF) [262] model improved
to some extent the existing MTT [254,264,266–269] by the introduction of solvent-shared specific counterion-headgroup
pairs. However, the MF model only applies to the spherical and globular micelles in a very narrow range of added salt con-
centration and does not take into account the sphere-to-rod transition and growth of micelles to long cylinders. Koroleva and
Victorov (KV) [135] developed a model that introduces the specific headgroup-counterion pair in which a geometric param-
eter, called the distance of the closest approach of the ion to the core, is added to take into account hydration effects. More-
over, they adopted a modified Poisson-Boltzmann (PB) equation [270] that incorporates the dispersion interactions between
ions and micelles to differentiate the polarizability of different ions, which has also been considered by recently reported
work [271]. The predicted cmc, aggregation number, and sphere-to-rod transition are in reasonably good agreement with
experiment. However, the incorporation of the dispersion interaction between counterion and micelle in the modified PB
equation does not adequately reflect the effect of counterion specificity on cmc [262].
It is reported that an alternative molecular thermodynamic (AMT) [9,136,249,250] model for the prediction of the aque-
ous cmc based on existing MTT is developed which incorporates the surfactant activity, counterion activity, and ion effects
on surfactant aggregation. In the developed model, the activity coefficient of ions/counterions is evaluated using Pitzer’s
method [272,273] or Davies [274] equation depending on the salt concentration, the activity coefficient of surfactants’
hydrocarbon tail is evaluated from the Setchenov equation [275], and the specificity of headgroup-counterion pair is consid-
ered to reflect hydration effects and the degree of counterions binding to micelles. The counterion binding coefficient is ini-
tially set as a variable and finds its optimal value by minimizing micellization free energy. The effect of coion is evaluated
from salt-dependent factors [9,136,249,250], including the Setchenov coefficient, the dielectric decrement of salt, and the
correlation between the change of surface tension and the change of salt concentration in aqueous solution.
The aqueous cmc of pure surfactant i (Cw i Þ, or of surfactant mixture (C ), is calculated using the equation below (C is
w w
used for illustration) [9,136,249,250]:
Cw ¼ ðC mw þ C s ÞexpðkT Dlm Þ
1 o
ð100Þ
where C mw is molar concentration of water, C s is concentration of salt, k is the Boltzmann constant, T is temperature, and
Dlom is micellization free energy which is calculated from several contributing thermodynamic terms [9,136,249,250]:
Dlom ¼ Dlotrt þ Dloint þ Dlopack þ Dlost þ Dloent þ Dloelec þ Dloact ð101Þ
The first three terms on the right side of Eq. (101) are associated with the packing and interactions of hydrocarbon tails
and the formation of a hydrophobic micellar core: Dlotrt , Dloint , and Dlopack represent free energy contributions from hydro-
carbon transfer from water into micelle, formation of micellar core-water interface, and hydrocarbon tail packing in micelle,
respectively. The next three terms are associated with surfactant headgroups and counterions in the micelle-water interfa-
cial region: Dlost , Dloent , and Dloelec represent surfactant headgroup steric interactions, headgroup-counterion mixing, and
electrostatic interactions, respectively. The last term Dloact represents the contribution from surfactant activity and counte-
rion activity in the bulk solution [9,136,249,250].
The application of this model has been validated using various pure and mixed surfactants in salt solutions
[9,94,95,136,249,250]. The model application to pure alkyltrimethylammonium surfactant CnTABr in solution with added
salt (NaBr, NaCl, or KCl) to evaluate chain length effects, counterion effects, and coion effects on aggregation properties is
shown in Fig. 29 [136]. The aqueous cmc (Fig. 29(a)) and the sphere-to-rod transition threshold (Fig. 29(b)) decreases as
chain length increases whereas the weight-based aggregation number Nw (Fig. 29(b)) increases as chain length increases.
The predicted aqueous cmc for all surfactants in Fig. 29 match very well with experimental data except that a slight devi-
ation appears for C12TABr with added NaBr above 1 M. An excellent agreement is observed between predicted and experi-
mental Nw. The sphere-to-rod transition is manifested by the sharp change of aggregation number, counterion binding
coefficient, and core minor radius (not shown here) as a function of salt concentration. The comparison of model predicted
transition (salt concentration threshold) and deduced transition from experiment [136,250,276–283] match reasonably well.
For C16TABr with added salt KBr, for example, the predicted threshold is 0.08 M and the experimental threshold is 0.1 M
[281–283].
The Hofmeister series, which indicates Cl < Br for cationic surfactant aggregation, is reflected by the effect of counterion
on the depression of the aqueous cmc and of counterion binding coefficient, and on the increment of Nw by comparing
C16TABr and C16TACl (see Fig. 29) [136]. The effect of coion is examined by adding different salts (NaBr and KBr) to the
aqueous solutions containing C16TABr: the coion effect on cmc and on Nw is minor at low salt concentration, whereas the
effect of increasing salt concentration, which increases the coion effect, becomes increasingly noticeable, as shown in
Fig. 29 [136].
Fig. 29. Comparison of predicted parameters and experimental parameters of aggregation: (a) cmc, (b) weight-based aggregation number Nw, and (c)
counterion binding coefficient of CnTAX (X = Br, Cl) versus salt concentration. The salt type is specified in the plots; otherwise, the salt is defaulted to
NaBr. Solid and dashed lines represent model prediction; symbols represent experimental data cited from references. Model inputs are based on
experimental conditions: total concentration of surfactant set at 10 mM for C14TABr and C16TABr/Cl, and at 30 mM for C12TABr at temperature of 35 °C.
(After Ref. [94,136]).
It is reported that the AMT model can also be applied to ternary surfactant mixtures, such as cationic/cationic/nonionic
mixture: C16TABr/C16BzCl/C16E20 with added NaCl in the aqueous solution, as shown in Fig. 30. Fig. 30(a) gives the
comparison of predicted cmc from various models, including the Clint model [284], Rubingh and Holland (R-H) model
[285,286], and the AMT model [9,136,249,250], and experimental cmc data [287]. The AMT model gives the best prediction
while there is an overestimation from the Clint model and an underestimation from R-H model. The predicted aggregation
number is calculated only from the AMT model, which gives slightly overestimated but reasonable values.
An improved traditional model [89,91] for the work of a similar kind [87,88] for the prediction of the aqueous cmc is also
reported and given below for various pure, binary, ternary, and multiple homologous/nonhomologous surfactant mixtures:
1
Cw ¼ ð1þdi Þ ð102Þ
X
xi ðcc C c Þdi C1p
i
i
where xi is the bulk mixed molar fraction of surfactant i. Cpi is the aqueous cmc of surfactant i in pure water (i represents
surfactant 1, 2, or 3, . . .). di is counterion binding coefficient with respect to surfactant i based on best-fit of experimental
data. di is nearly constant for a series of homologous surfactants and is also constant as a function of salt concentration
(low to medium depending on specific surfactant class: 0–1) [89,91]. Note that the counterion binding coefficient dj in
the advanced cmc model is with respect to counterion j and different from di. Cc is the concentration of ion dissociated from
electrolyte and from ionic surfactant in aqueous solution. cc is the mean activity coefficient of ions in aqueous solution and is
usually calculated using Pitzer’s method [272,273] or Davies [274] equation. Eq. (102) is supported by the report that the
cmc is heavily dependent on and exponentially related to electrolyte concentration [251,252,271,288]. The application of
this improved traditional model for the aqueous cmc prediction is shown in Fig. 31 [91] and details of cited experimental
data can be found elsewhere [96,100,287,289–292].
It is clear that the aqueous cmc, which is usually predicted with existing models [9,89,91,136,249,250], takes into account
the ion/salt effect on aggregation/adsorption, headgroup-counterion pair and associated hydration effect, hydrocarbon chain
length, van der Waals interactions between surfactant molecules, steric interactions between head groups, electrostatic
Fig. 30. Predicted (a) cmc, and (b) aggregation number of ternary mixed surfactants alkyltrimethylammonium bromide C16TABr, benzyl dimethyl
hexadecyl ammonium chloride C16BzCl, and polyoxyethylene (20) cetyl ether C16E20 versus experimental results. Predicted values in (b) were calculated
using AMT model. Inputs of model according to experiment conditions: total surfactant concentration set at cmc in 30 mM NaCl solution at 25 °C. (After Ref.
[136]).
Fig. 31. Comparison of predicted cmc and experimental cmc: (a) the cmc of pure CnTAB as a function of HCl concentration in solution at 25 °C; (b) the cmc
of binary-mixed nonionic surfactants (C9H19KO2 and C11H23KO2) as a function of molar fraction of C9H19KO2 in bulk solution at 25 °C; (c) the cmc of binary-
mixed anionic and nonionic surfactants (SDS and OG) as a function of molar fraction of OG in bulk solution containing 20 mM NaCl at 25 °C; (d) the cmc of
binary-mixed nonionic surfactants (C9COOE12 and C11COOE12) as a function of molar fraction of C9COOE12 in bulk solution at 25 °C; (e) the cmc of ternary
mixed homologous cationic surfactants BAC (C12, C14, & C16) as a function of molar fraction of C14 in bulk solution containing 0.0342 M, 0.171 M, and
0.856 M NaCl at 40 °C; C12 & C16 are equal in molar fraction; (f) the predicted cmc compared to the experimental cmc of ternary mixed cationic, cationic, and
nonionic surfactants (C16, C16TAB, and C16E20) at various bulk mixed molar ratios in solution containing 30 mM NaCl at 25 °C. Symbols represent
experimental data; lines represent model predicted data. CnTAB: n-alkyl trimethyl ammonium bromide; CnH(2n+1)KO2: potassium alkanoate; OG:
octylglucoside; SDS: sodium dodecyl sulfate; CnCOOE12: CnH(2n+1)COO(CH2CH2O)12CH3; Cn: n-benzalkonium chloride or BAC; C16E20: polyoxyethylene cetyl
ether. (After Ref. [91]).
interactions at the interfacial region of micelles, and the interactions between solvent and surfactant. Therefore, the insertion
of the aqueous cmc into the Langmuir isotherm, which is the MLA as introduced previously in Eq. (81), can accurately describe
the adsorption phenomena and adsorption of surfactants on substrates and associated effects of physical and chemical prop-
erties of surfactants and solvent environments. Beyond the applicability of the MLA model for pure surfactant and mixed
homologous surfactants, a valuable part lies in its potential to evaluate the corrosion inhibition of various surfactant mixtures
of different classes under various solution conditions using only one set of fit experimental data. Examples are given in Fig. 32
with the assumption that the corrosion inhibition efficiency is equal to the effective metal surface coverage [185].
7.3. Water/oil partitioning
When an aqueous surfactant solution comes into contact with an immiscible organic liquid, such as oil, surfactant mono-
mers may prefer partitioning into organic liquid until equilibrium is reached between the two liquids [95,181,293–295].
Considering the complicity of water/oil partitioning of surfactants in water-oil environment and associated interfacial phe-
nomena, the determination of surfactant partitioning between water and oil usually serves as the basis of the hydrophobic-
hydrophilic balance [95,181,294–297], which further affects the availability of monomeric surfactants in aqueous phase and
the associated adsorption on metal surface for corrosion inhibition [95,181].
For pure surfactant, the partitioning is usually characterized by the partitioning coefficient, which is defined as the ratio of
monomeric surfactant concentration in oil to that in aqueous phase (pure water or salt-containing water) [95,298]:
C omi
Ki ¼ ð103Þ
Cw mi
where K i is the partitioning coefficient of surfactant ‘i’, C omi and C w

mi are monomeric concentration of surfactant ‘i’ in oil phase
and aqueous phase, respectively.
Extensive research has been performed on low concentration (typically lower than the aqueous cmc) partitioning of non-
ionic surfactants [236,295–306]. The partitioning research on higher surfactant concentration systems, however, has been
rarely reported and limited [95,307,236,308,309]. The relevant report on the partitioning of ionic surfactant at high concen-
tration levels is even less than that for nonionic surfactants [95,181,310]. The investigation of partitioning above the aqueous
Fig. 32. Comparison between the experimental and the predicted inhibition efficiency of different testing systems of mixed BAC-X65 steel at 40 °C and of
the reported testing systems containing pure surfactants AAOA and C16TAB, and surfactant mixture TCA-DDPB. All symbols represents experimental data;
lines represent MLA predicted data; the solid line represents LA model predicted data for comparison. K0 are 15.73 for AAOA-1018 steel system in solution
containing 0.856 M NaCl at 25 °C, 2.52 for C16TAB-copper system in solution containing 0.03 M Fe(NO3)3 at 32 °C, and 20.26 for TCA-DDPB-J55 steel in
solution containing 10% HCl at 30 °C. Testing Systems I and II are mixtures of BAC surfactants (C12, C14, and C16) under different experimental conditions.
AAOA: N-[2-[(2-aminoethyl) amino] ethyl]-9-octadecenamide; TCA: trans-cinnamaldehyde; DDPB: dodecylpyridinium bromide. (After Ref. [185]).
cmc and the apparent cmc is important (the apparent cmc is the average concentration in water and oil environment at
which the micelle starts to form): the portioning is a monomer process, and the partitioning coefficient is determined by
monomer concentrations in the two phases, which are limited by micelle formation [95,181].
For surfactant mixtures, the partitioning becomes more complicated in terms of equilibrium mixture composition in each
phase, because of the effect of individual mixed species on the partitioning, and the adsorption of mixture at the oil/water
interface [95,181]. It has been shown that for some pure surfactants, a plateau concentration of monomer is reached either in
oil phase or in aqueous phase with increasing total surfactant concentration beyond the aqueous cmc [307,311,312]. How-
ever, it is also reported for mixed surfactants that the amount of surfactants partitioned into the oil phase continues to
increase beyond the aqueous cmc [236,309,311]. The partitioning change of mixed surfactants above the aqueous cmc is
reported to arise from the selective partitioning of more hydrophobic components into oil phase, which makes the experi-
mental investigation and quantitative modeling work more challenging [236,307,311].
Before moving onto the discussion of partitioning modeling, it is necessary to clarify the relation between partitioning
and the aqueous cmc. The aqueous cmc of pure surfactant or mixed surfactants in the absence of oil phase is assumed to
be equal to the aqueous cmc in the presence of nonpolar oil phase, which is confirmed by related reports [95,181]. On
the other hand, the nonpolar oil phase does not contribute to the micelle formation in aqueous phase. It is actually reported
that for nonionic surfactants with non-polar heptane as oil phase, the aqueous cmc has been observed to be very similar to
the corresponding aqueous cmc without oil phase [313] and that for certain anionic surfactants with heptane as the oil
phase, the aqueous cmc has also been found to be very close to the cmc measured in water in the absence of oil [314].
For certain cationic surfactants with a more polar oil phase (dichloromethane), however, the aqueous cmc in the absence
of oil phase is significantly different from the corresponding cmc in the presence of oil phase [165].
It has been reported that the partition coefficients of surfactant in pure water/oil environment can be predicted using
semi-empirical modeling [296,304] and quantum chemical methods [104,315,316]. One quantum prediction of partitioning
coefficient has been reported to take into account the effect of protonation in aqueous phase [317,318], which is, however,
far away from realistic conditions in oilfields where the aqueous phase contains multiple classes of inorganic salts, and the
crude oils are complex mixtures of organic solvents. It is a challenging task to estimate surfactant partitioning between salt-
containing aqueous phase and the phase of organic mixtures using the partition coefficient for pure-water/single-oil system
despite the availability of developed theories which are at different stages of maturity [319–325].
Very recently, an improved surfactant partitioning prediction model termed water/oil surfactant distribution model has
been reported for the evaluation of partitioning and distribution of mixed surfactants in water (containing salts) and oil envi-
ronment [95,181]. With this model, the partitioning coefficient Ki of surfactant ‘i’ is predicted using the following equation
[95]

cwmi C mo Dlotri
Ki ¼ exp ð104Þ
comi C mw RT
where comi and cw o

mi are activity coefficients of monomeric surfactant ‘i’ in oil phase and water phase, respectively. cmi is usually
assumed to be unity. Details of calculation of cw mi have been reported elsewhere [95], in which different calculation methods
for ionic surfactant and nonionic surfactant were introduced. For ionic surfactant, cw mi is equal to the geometric mean of the
activity coefficient of counterion and the activity coefficient of hydrocarbon tail, whereas for nonionic surfactant, cw mi is equal
to the activity coefficient of hydrocarbon tail. The activity coefficient of ions/counterions is evaluated using Pitzer’s method
[272,273] or Davies [274] equation depending on the salt concentration, whereas the activity coefficient of hydrocarbon tail
is evaluated from the Setchenov equation [275]. The essence of cw mi is to take into account the effect of dissolved salt in water
on water-oil partitioning of surfactants. C mo and C mw are molar concentrations of oil and water, respectively. The standard
free energy change of transfer of surfactant ‘i’, Dlotri , from water to oil is estimated from two methods. Method I is the free
energy transfer method [95], in which Dlotri is calculated as the sum of the headgroup transfer energy and the hydrocarbon
tail transfer energy. Method II is the quantum chemical method [104], in which Dlotri is interpreted as the difference in sol-
vation energy of surfactant ‘i’ in oil and in water based on the quantum chemical calculations using simulation software,
such as Gasussian09. An excellent agreement is observed between predicted and experimental values of Ki for various sur-
factants as shown in Fig. 33 [95].
The comparison of predicted and experimental partitioning coefficients of pure BAC surfactants (C12, C14, and C16) in
water (salt containing) – oil environment is shown in Fig. 33(a). The predicted partitioning coefficients at various conditions
based on the transfer free energy calculated using the two aforementioned methods match the experimental data reasonably
well. Note that the transfer free energy of the polar functional group of BAC surfactants is 55.3 kJ/mol at 40 °C [95]. This value
is much lower than the reported value of the polar functional group (>N(CH3)+2) transferring from 0.1 M aqueous sodium
hydroxide to heptane [326]. This is probably because the functional group in BAC surfactants has one extra benzene group
and one extra methylene group, which prefer the non-polar organic phase and thus decrease the free energy of transfer. The
affinity of the functional group to different organic phases also contributes to the energy difference.
The method to determine the transfer free energy for the partitioning process was also tested on other surfactants. Fig. 33
(b) presents the predicted and experimental partitioning coefficients of homologous polyoxyethylene glycol n-dodecyl ether
C12H25(OCH2CH2)nOH (or C12En) surfactants in pure water and isooctane environments; Fig. 33(c) presents the predicted and
experimental partitioning coefficients of N-based alkyl amines and derivatives in 0.1 M NaOH water and heptane environ-
ments [95]. As can be seen, there is an excellent agreement between predicted and experimental partitioning coefficients.
Assuming the amount of adsorbed surfactants in the oil-water interface is negligible a mass balance of total mixed sur-
factants in the water-oil environment is given by [95,181]
C tol V w ¼ CðV w þ V o Þ ð105Þ
where C tol is the initial concentration (not at equilibrium) of total surfactants added to aqueous phase. C is the overall aver-
age concentration of total surfactants in water-oil environments. Vo and Vw are volumes of oil and water, respectively.
When C < Cap (Cap is the apparent cmc which is defined as the average concentration of mixed surfactants at which mixed
micelles start to form in water-oil environments), mass balance of each mixed surfactant ‘i’ at partitioning equilibrium is
given by [95]
xi C tol V w ¼ C w o
mi V w þ C mi V o ð106Þ
Fig. 33. Comparison of the predicted partitioning coefficients and the experimental partitioning coefficients: (a) the partitioning coefficients of pure C12,
C14, and C16 between water and oil (toluene) at 40 °C. [For symbols in (a), open symbols with vertical center-cross line: transfer free energy calculated using
Method II; all other symbols: transfer free energy calculated using Method I. Open symbols: 0 M in NaCl water; open symbols with center dot: 0.0342 M NaCl in
water; open symbols with (vertical and horizontal) center-cross line: 0.171 M NaCl in water; half-filled symbol: 0.804 M NaCl in water; solid-filled symbols: 0.856 M
NaCl in water.] (b) The partitioning coefficients of polyoxythylene alkyl ether (C12En) in pure water and isooctane environment at 25 °C. (c) The partitioning
coefficients of alkyl amines in 0.1 M NaOH water and heptane at 20 °C. (After Ref. [95]).
The partitioning coefficient of surfactant mixture is termed the apparent partitioning coefficient and is given by [95]
P o
C om C
K mix ¼ ¼ P mi ð107Þ
Cw m Cwmi
where C om and C w
m are total concentrations of monomeric surfactants in oil phase and aqueous phases, respectively.
Substitution of Eqs. (103) and (106) into Eq. (107) leads to [95]
P
K i xi =ðV w þ V o K i Þ
K mix ¼ P ð108Þ
xi =ðV w þ V o K i Þ
where xi is the molar fraction of surfactant ‘i’ in total mixed surfactants in bulk solution. Eqs. (106) and (108) are only appli-
cable to the condition of C < Cap , whereas Eqs. (103) and (107) are applicable to all values of C.
When C > Cap , it is assumed that partitioning process between water and oil only involves monomers. For ionic surfactant,
the partitioning involves the surfactant molecule and the associated counterion. On the other hand, there is no dissociation
in the process of partitioning. C omi , C w w
mi , and C m are given by [95,181]
mi ¼ f i ai Ci
Cw w
ð109Þ
C omi ¼ f i ai K i Cw
i ð110Þ
X
Cw
m ¼ Cw
mi ð111Þ
f i is the activity coefficient of surfactant ‘i’ in a micelle, and it is assumed to be unity. ai is the molar fraction of surfactant ‘i’ in
mixed micelles and is given by [95]
xi C tol
ai ¼ ð112Þ
C tol C m ð1 þ V o =V w Þ þ f i Cw
i þ f i K i Ci V o =V w
w
where C m is the average concentration of total monomeric surfactants in water-oil environments. Summation of the molar
fraction of surfactant ‘i’ in the mixed micelle should be unity and thus [95]
X xi C tol
¼1 ð113Þ
C tol C m ð1 þ V o =V w Þ þ f i Cw
i þ f i K i Ci V o =V w
w
For fixed values of other parameters, Eq. (113) is a polynomial function of Cm. For surfactant mixtures with multiple com-
ponents, Cm has multiple corresponding mathematical values. However, in reality Cm should only have one value and should
be confined in the range of [95]
Cap < C m < C ð114Þ

Eqs. (113) and (114) are solved simultaneously with respect to Cm using a home-designed MATLAB program. The appar-
ent cmc of surfactant mixture in water-oil environments is given by [95]
1
Cap ¼ P xi ð115Þ
f i Cw
i
V w =ðV w þV o Þþf i K i Cw
i
V o =ðV w þV o Þ
The apparent cmc for pure surfactant ‘i’ in water-oil environments is given by [95]
K i Cwi V o þ Ci V w
w
Cap
i ¼ ð116Þ
xi ðV o þ V w Þ
This defined apparent cmc of one pure surfactant in water-oil environments can reflect the relative hydrophobicity/
hydrophilicity of that surfactant. The higher the apparent cmc of one pure surfactant, the lower the hydrophobicity of that
surfactant in water-oil environments. The reverse is also true.
With this developed water/oil surfactant distribution model, the partitioning coefficient Ki of surfactant i, the aqueous
cmc of surfactant i, the apparent cmc of mixed surfactants in water-oil environment, Cap , monomer concentration of surfac-
mi and C mi , and molar fraction of surfactant i in the mixed micelles, ai, can be predicted in
tant i in water and in oil phases, C w o
water (containing salts)-oil (nonpolar) at given inputs which include total surfactant concentration and mixed molar ratio xi
in bulk solution. If experimental data for the cmc and partitioning coefficient of surfactant i are available, it is best to use the
experimental data. However, if no experimental data are available, it is best to use the methods introduced above to predict
the aqueous cmc and the partitioning coefficient and then substitute these values into the surfactant distribution model.
The application of water/oil surfactant distribution model to the water-oil partitioning of equal-molar ternary mixtures of
BAC surfactants is shown in Fig. 34 [95]. Fig. 34(a) and (b) presents equilibrium concentrations of monomeric surfactants in
water and in oil versus total initial concentration of mixed surfactants added to the aqueous phase. The intersection of the
vertical dashed line and horizontal axis identifies the aqueous cmc of surfactant mixture, which is Cw = 3.40 105 M. As
can be seen, the partitioning of each mixed component as well as the overall partitioning continues without any change
Fig. 34. Properties at equilibrium partitioning of equal-molar mixed BAC surfactants (C12, C14, and C16) in water (0.171 M NaCl)-oil environment at 40 °C:
(a) concentration of monomeric surfactants in aqueous phase, (b) in oil, (c) concentration of total surfactants in aqueous phase, including monomeric and
micellized form, and (d) micelle composition of surfactant ‘i’ as a function of initial concentration of total surfactants added to aqueous phase. Symbols:
experiment; lines: model prediction. Vertical dash line represents the cmc of surfactant mixture in aqueous phase: Cw; vertical dot line represents twice of
the apparent cmc of surfactant mixture in water-oil environment: 2Cap . (After Ref. [95]).
when Ctol reaches Cw. This is consistent with the reported view that the amount of surfactants partitioned into the oil phase
continues to increase beyond the aqueous cmc of that mixture [236]. It is easy to understand that the partitioning of surfac-
tants into the oil phase depletes the surfactants in the aqueous phase to an extent that prevents micelle formation in the aque-
ous phase at Ctol < 2Cap. 2Cap is used as the upper limit rather than Cap considering that water and oil are equal-volume mixed
and that the horizontal axis represents the total initial concentration of surfactants, Ctol, added to the aqueous phase [95].
Above 2Cap, which is suggested by the dotted line, the partitioning behavior of each mixed surfactant component starts to
change as indicated by the transition point in Fig. 34(a)–(c) [95]. The mixed surfactants form micelles in aqueous phase as
indicated by Fig. 34(c). It is also interesting to observe that above the transition point, monomeric concentrations of C14 and
C16 in both phases decrease while the concentrations of more hydrophilic C12 in both phases continue to increase to some
extent, and that the total monomeric concentration in aqueous phase (C w o
m Þ increases and the counterpart in oil (C m Þ phase
decreases slightly before reaching a plateau. This phenomenon can be explained by the fact that surfactant molecules in
micellar form are generally more thermodynamically stable than in monomeric form with respect to relatively hydrophobic
species and that C16 and C14 prefer to exist in micellar form, which leads to the leveling off of monomeric surfactant concen-
tration in water and oil phases. The preference of the micellar form of C16 and C14 is reflected by the much higher molar frac-
tion in micelles at the beginning of micelle formation as shown in Fig. 34(d) [95], indicating the formation of more
hydrophobic micelles at the beginning. As the total surfactant concentration increases, the micelles become less hydrophobic.
The relative hydrophobicity/hydrophilicity of one particular surfactant relative to the other in a one-phase system, such
as in water phase, may or may not be the same as that in a two-phase system, such as in the water-oil phases. The relative
hydrophobicity/hydrophilicity of surfactant can usually be reflected by the cmc value in the environment discussed, which is
termed the apparent cmc, as defined for pure surfactant ‘i’ in water-oil environment in Eq. (116). For data from mixed BAC
surfactants in water-oil environments shown in Fig. 34, the apparent cmc of each mixed component is calculated based on
the aqueous cmc and partitioning coefficient, and the relative values are given below [95]
Cap
C12 > CC14 > CC16
ap ap
ð117Þ
Therefore, C16 is still the most hydrophobic surfactant among the three in water-oil environments, as well as in water
only. It is thus easy to understand that C16 has the highest mixed molar fraction at the beginning of micelle formation in
water-oil environments, followed by the mixed molar fractions of C14 and C12. As Ctol increases, the molar fraction ai in mixed
micelles of each mixed component approaches the initial mixed molar ratio xi in the bulk solution. The relative values of Cap i
also shed light on the continuing increase of the concentration of monomeric C12 and the decrease of that of C14 and C16 after
micelles formation as shown in Fig. 34(a) and (b).
The application of a water/oil surfactant distribution model is also extended to water-oil partitioning of additional sur-
factants, as shown in Fig. 35(a) and (b) for mixed primary alcohol ethoxylates C12OE14 and C12OE30, Fig. 35(c) and (d) for
mixed hexaoxyethylene nonyl phenyl ether (NPE6) and octaoxyethylene nonyl phenyl ether (NPE8), and Fig. 35(e) and (f)
for mixed C16 and polyoxyethylene 20 cetyl ether (C16E20) in water-oil environments [95]. As can be seen from Fig. 35(a),
Fig. 35. Comparison between predicted and experimental partitioning properties of surfactants. Figures (a) and (b) are equilibrium partitioning properties
of mixed C12OE14 and C12OE30 surfactants in water-oil (trichloroethylene) environment at 25 °C: (a) concentration of surfactants, (b) average ethoxylate
group (EO) distribution in aqueous and oil phase as a function of equilibrium aqueous concentration Cw. The values of aqueous cmc are 123.2 mg/L and
560 mg/L for C12OE14 and C12OE30, respectively. Mixed ratio: 0.475/0.525. The arrow in (b) indicates the initial EO average in water-oil environment. Figures
(c) and (d) are equilibrium partitioning properties of mixed NPE6 and NPE8 surfactants in water-oil (cyclohexane) environment at 25 °C: (c) concentration of
monomeric surfactants in oil phase (d) molar fraction of surfactants in mixed micelles as a function of Ctol. The values of aqueous cmc and partitioning
coefficients are 2.70 105 M and 4.05 105 M, and 481 and 70 for NPE6 and NPE8, respectively. Mixed ratio: 0.542/0.458. Figures (e) and (f) are
equilibrium partitioning properties of mixed C16 and C16E20 surfactants in water-oil (heptane) environment at 25 °C: (e) concentration of monomeric
surfactants in 0.03 M NaCl aqueous phase (f) in oil phase as a function of Ctol. The predicted values of aqueous cmc and partitioning coefficients from
previous work are 3.61 105 M and 2.47 106 M, and 5.32 and 0.66 for C16 and C16E20, respectively. Mixed ratio: 0.542/0.458. Lines: model prediction;
symbols: reported data. (After Ref. [95]).
the predicted and experimental [236,307] surfactant distribution as well as the ethoxylate group (EO) average per molecule
in oil phase (trichloroethylene) for mixed C12OE14 and C12OE30 agree reasonably well.
However, the experimental EO average is slightly higher in the aqueous phase probably due to experimental error con-
sidering that the EO average should never be higher than 30 [307]. The partitioning into the oil phase continues above the
reported aqueous cmc value of 240 mg/L for the mixture [308], which is mainly because of the selective partitioning of more
hydrophobic component C12OE14 into the oil phase. This is also accompanied by an EO average increase in the aqueous phase
and decrease in the oil phase relative to the initial EO average as indicated by the arrow in Fig. 35(b). An excellent agreement
between the predicted and reported [311,327] monomeric surfactant distribution and molar fraction in micelles is observed
in Fig. 35(c) and (d) for mixed NPE6 and NPE8 in equal volume water/oil (cyclohexane) environment. When Ctol reaches 2Cap
(Cap = 2.5 103 M), more hydrophilic micelles with a higher molar fraction of NPE8 start to form, which seems to be con-
trary to the initial formation of more hydrophobic micelles of mixed BAC surfactants in the water-oil environments. By
examining the apparent cmc of pure NPE6 and NPE8, however, it is found that NPE8 has a lower apparent cmc
(2.89 103 M) than NPE6 (1.30 102 M) in water-oil environment. Therefore, it is expected that NPE8 should have a
higher molar fraction in mixed micelles at the beginning of micelle formation. Similar interpretation is applicable to mixed
C12OE14 and C12OE30 in water-oil environment.
It is interesting to note that when the surfactant distribution model is applied to non-homologous mixed C16 and C16E20 in
water-oil (heptane) environment, good agreement between experiment and model prediction is observed in Fig. 35(e) and (f)
with respect to the monomer distribution in each phase [95]. The aqueous phase cmc and partitioning coefficient are pre-
dicted from the developed cmc model and the partitioning coefficient calculation method as described previously. It is
believed that this is the first time that the partitioning and distribution of non-homologous mixed surfactants in oil-
water environments have been evaluated both theoretically and experimentally, which is important to industrial production
processes in terms of contamination control and investment minimization.
This developed water-oil surfactant distribution model for the evaluation of the water-oil partitioning of surfactant is
proposed and validated. This model is applicable over a wide total surfactant concentration range, including the aqueous
cmc and the apparent cmc. The model predicted data and the experimental/reported data of surfactant distribution in
water-oil environments agree very well. However, the application of this model should be limited to nonpolar or slightly
polar organic solvents that do not affect the aqueous cmc significantly as well as to systems in which microemulsion forma-
tion is avoided. These limitations should be addressed in future modeling work [95].
7.4. Surfactant precipitate and colloid formation
Ionic surfactants usually interact with other ions in solution. As surfactants encounter reactive counter ions they can react
to form precipitates. Some surfactants such as carboxylates react with hydrogen to form carboxylic acid, which is often in the
form of oil droplets. Carboxylates can also react with other positively charged species such as calcium or magnesium ions to
form solid, colloidal, hydrophobic particles that behave like microscopic pieces of wax. The process of precipitation to form
neutral molecules greatly reduces water solubility. However, the hydrophobic colloidal precipitate particles are often soluble
in nonpolar solvents such as oils. Thus, precipitation has a very pronounced effect on phase partitioning that strongly favors
the lipophilic or oil phase.
Ionic surfactants interact with hydrogen, hydroxide, and other ions. Consequently, phase stability is often a function of pH
[328]. Fig. 36 shows the stability regions for oleic acid/sodium oleate in a solution containing calcium ions as noted. Oleate
-2
oleate
-3 micelles
Log(Oleate Concentration mol/l)
-4 oleic
calcium
acid
dioleate
-5
-6
-7 oleate
-8 dissolved
oleic acid
-9
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
pH
Fig. 36. Comparison of species with respect to total oleate concentration and pH in the presence of 0.0001 M calcium ions. (After Ref. [328]).
reacts with hydrogen ions to form oleic acid at low pH. The diagram also shows that at high pH and oleate concentrations,
aggregate structures form.
It has been reported that [104] the ions Fe2+ and Fe3+ from iron dissolution can combine with surfactant molecules to form
complexes or ligands which affect the availability of monomeric surfactants in bulk solution and thus compromise adsorp-
tion on metal surface and associated corrosion inhibition. It is also likely that the surfactant adsorption consists of such
ligands or complexes that affect the packing efficiency of monolayer/multiplayers. Other components, such as sand, can also
compromise the efficiency of surfactant inhibitors in a way that these components can act as an alternative adsorption sink
for the surfactant inhibitors. Besides the characterization of the specific complex formation processes using experimental
techniques, mechanistic models, which are usually based on a combination of the best-fit of experimental data and associ-
ated theory, may become more useful in a way that the developed model can be extrapolated to similar testing systems, such
as the previously mentioned MLA [89,91]. Please note that the quantum chemical methods might be used to evaluate com-
plex formation [104,329–332], but it is difficult to simulate the conditions in realistic WOS environments.
7.5. Fluid flow
It is necessary to take into account the flowing water/oil fluid in WOS environments for the evaluation of surfactant cor-
rosion inhibition efficiency because the inhibitor concentration profile changes over time and thus affects corrosion inhibi-
tion. The flow rate can be simulated using an experimental setup, such as a flow loop [333] or simply a rotating disk
electrode [89,91,334], multiphysics modeling, such as finite element modeling [335], or the combined mechanistic modeling
with experimental evaluation [89,91], so that a correlation between the flow rate and surfactant injection frequency can be
set up for the effective corrosion inhibition. Another challenge is that the vigorous flow may cause formation of microemul-
sions in either aqueous or oil phase or both, which are difficult to evaluate through modeling.
7.6. Salt/ion effects
The aqueous phase in oilfields generally contains mixtures of various inorganic salts, which not only promote the corro-
sion of metal in ways as discussed previously but also affect surfactant-associated processes, including aggregation, adsorp-
tion, partitioning, surfactant-ion pair, hydration, and thus affect corrosion inhibition. Any experimental evaluation and
modeling work should take this into account. Alternatively, the ion effects on the efficiency of surfactant adsorption may
be incorporated into certain processes associated with surfactants, such as aggregation and micellization which are well
accounted for by the above mentioned MLA [89,91,94,95,181] and water/oil surfactant distribution model [94,95,181]. At
present, these mechanistic modeling methods are well developed to describe the effect of simple 1:1 salts (such as NaCl).
More complicated salts (such as Fe2(SO4)3) will require additional work so that the model can be tuned for application in
more complicated systems with mixtures of salts.
7.7. Microstructure of metal
In the evaluation of corrosion inhibition efficiency of surfactants, it is recommended that for experiments, the metal sur-
face conditions should be consistent and for modeling work, certain thermodynamic and kinetics parameters should be con-
sidered to describe the metal surface state and microstructure evaluation as a function of time. The cathodic and anodic
reactions may be changed in ways that the state of metal surface is inevitably affected by the corrosive media, such as
pH, dissolved salts, and temperature and that the metal surface is generally inhomogeneous and contains defects and con-
taminants which preferentially initiate local corrosion attack ahead of uniform metal corrosion. Corrosion inhibition may be
affected by surfactant adsorption and preferential adhesion to certain microstructural features which promotes either the
cathodic or anodic reaction. For example, the interaction between surfactants and metal surface defects may alter the steady
state of protective oxide film which defines the long-term corrosion inhibition [336–338].
To minimize the error introduced by microstructure of metal surface, it is recommended all tests follow the same proce-
dures for sample surface preparation and cleaning and that try to avoid the preferential corrosion attack introduced by these
procedures [339–345]: for example, minimization of the edge effects when samples are cut from bulk metal and avoidance
of deep scratches when metal surface is polished. Furthermore, the degreasing agent should not be corrosive to metal sam-
ples, and cleaning of corrosion products should follow standard methods such as ASTM G1 [342], etc. More details about the
role of metal surface and microstructure and surface preparation in the corrosion inhibition have been reported elsewhere
[104,141].
Based on the previous discussion, it is clear that the evaluation of surfactant performance in WOS environment is complex
and that experimental characterization and multiphysics modeling should incorporate at least most of the major processes,
such as adsorption, partitioning, aggregation, and salt effects. Other relatively unknown [346] factors such as the influence of
the oil on inhibition, should also be evaluated.
8. Corrosion inhibition experimental evaluations
A suite of laboratory tests is recommended and performed for specific applications of pure or mixed surfactant inhibitors
to evaluate the associated corrosion inhibition performance before deployment in actual oilfields [141,147,340]. The test
conditions in the laboratory should be the same or as close as possible to the actual conditions in the oil pipelines, such
as the composition of tested metallic material, the temperature, salt concentration, etc. Tests are usually performed in flow
loops which simulate a realistic oil pipe [333] or simply a rotating disk electrode in glass cell [89,334] with real or simulated
environment. Commonly used test techniques in corrosion inhibition evaluations include linear polarization resistance,
potentiodynamic scans, electrochemical impedance spectroscopy, electrochemical noise, and weight loss measurement
[89,91,240,347].
9. Corrosion inhibition modeling evaluations
Many different mathematical models have been developed for the evaluation of corrosion inhibition using surfactant
inhibitors in WOS environment, including semi-empirical models, mechanistic models, combined semi-empirical and mech-
anistic models, and multiphysics models, which will be reviewed in this section.
9.1. Semi-empirical models
Semi-empirical models are usually based on best-fits of the experimental data using parts of known theory. Before the use
of these models, model calibration should be performed with a sufficiently large and reliable experimental data set so that
they can be extrapolated to other testing systems with confidence. However, the calibration usually leads to the appearance
of certain constants with sound physical meaning but others constants as arbitrary best-fit parameters. It is, therefore, very
likely that extrapolation can lead to unreliable and sometimes physically unrealistic results because of lack of fundamental
basis for the best-fit parameters. Another issue with semi-empirical models is that they only focus on the effective surface
coverage and adsorption energy while ignore other aspects that are essential for evaluation of corrosion inhibition of surfac-
tants, such as salt effects, micelle formation, and packing efficiency, etc.
One model of this kind is the quantitative structure activity relation (QSAR) approach [93,170–174], which assumes that
corrosion inhibition efficiency can be evaluated as linear/nonlinear coupling of quantum chemical descriptors with experi-
mentally determined inhibitor efficiencies. The coupling based on best-fit of experimental data yields a set of regression
parameters which has no theoretical basis and can hardly be extrapolated to other surfactants or testing systems. Conse-
quently, the resulting data is restricted to a narrow range of parameters [94,186].
Other types of semi-empirical models may include various developed adsorption isotherms, the details of which can be
found in Sections 7.1.1.1–7.1.1.7.
9.2. Mechanistic models
Mechanistic models have a sound theoretical basis in ways that they can describe the mechanisms of corrosion inhibition
using surfactant. Most constants in these models have a clear physical meaning, despite the fact that the determination of
some constants still requires best-fits of limited experimental data. When established using one set or limited sets of reliable
experimental data, these models can produce accurate predictions with sound physical meaning, as well as reliable extrap-
olated predictions to other systems. Improvement of these models is feasible by adding parameters or coupling with other
approaches.
One such model is introduced in Section of ‘‘7.1.1.8”, in which the mechanistic MLA model is introduced [89,91,94,181].
Similarly, a modified QSAR model, the MQSAR model [94,136], is an improvement over most existing QSAR models in that
the aqueous cmc is coupled with the QSAR model to take into account various physical phenomena and solution
environments.
The MQSAR model can be described by Eq. (118) with respect to the effective surface coverage:
! !
ðA0 Q þB0 ÞC w
h¼ ! !
m
ð118Þ
Cw þ ðA0 Q þB0 ÞC w
m
! !
where A0 is a modified vector of regression coefficients, and B0 is a modified regression constant. Q is a vector of quantum
chemical descriptors for a particular surfactant or surfactant mixture. For the mixture of surfactants the quantum chemical
descriptors are weight-based average values of surfactant components.
The cmc, which is usually determined by rigorous thermodynamic models as mentioned above, actually takes into
account the various physical phenomena such as surfactant aggregation/adsorption, difference in chain length, and van
der Waals interactions between surfactant molecules, etc., and solution environments such as salt concentration and ionic
species [94,181]. Therefore, the insertion of the aqueous cmc into the regular Langmuir adsorption and QSAR can accurately
describe the adsorption phenomena of surfactants on substrates (metal electrode) and associated effects of physical and
chemical properties of surfactants and solvent environment. The set of regression parameters which is obtained from the
best-fit of experimental data of only one surfactant can be extrapolated to various pure and mixed homologous series of this
surfactant class and to various pure and mixed surfactants of similar classes. The more valuable part lies in its potential to
evaluate the corrosion inhibition of various surfactant mixtures of different classes at various solution conditions using only
one set of experimental data [94,181]. Examples of comparison of MLA- or MQSAR-based predictions and experimental data
are given in Fig. 37 for corrosion inhibition evaluation assuming corrosion inhibition is equal to effective surface coverage.
The experimental parameters of associated different testing systems are summarized in Table 4.
9.3. Multiphysics models
9.3.1. General description

The basic idea of a multiphysics model in the evaluation of the corrosion inhibition performance of surfactants in WOS
environments is that all the conceivable processes and phenomena that affect the overall surfactant corrosion inhibition
should be considered and incorporated into one integrated model in which connections between various processes and
resulting effects on ultimate corrosion inhibition is explored with submodels associated with critical processes that are inte-
grated and evaluated. Unknown factors that may affect surfactant performance are uncovered and incorporated, and one or
more software packages are utilized, based on the fundamentals from many areas of corrosion science, electrochemistry,
metallurgical engineering, and chemical and analytical engineering, etc. Existing models/submodels and computational
and programming resources are used for the description of associated processes. Other potential applications of such a
model may be extended to the design of surfactants, selection of optimal surfactants for specific applications, experimental
validation of developed models, and simulation of conceivable processes and phenomena. Such modeling can be integrated
into comprehensive lifetime prediction models in which all the parameters affecting corrosion inhibition may be evaluated.
Fig. 37. Comparison of MLA and MQSAR models predicted inhibition efficiency and experimental inhibition efficiency as functions of surfactant
concentration for various testing systems I-V. I: mixture of C12/C14/C16 = 0.70/0.25/0.05 in 0.171 M NaCl aqueous media with pH = 4 and electrode of X65
steel at 40 °C; II: mixture of C12/C14/C16 = 1/1/1 in 0.599 M NaCl aqueous media with pH = 5 and electrode of X65 steel at 40 °C; III: AAOA in 0.856 M NaCl
aqueous media with pH = 6 and electrode of 1018 steel at 25 °C; IV: cetylpyridinium chloride in 1 M HCl aqueous media with electrode of 1018 steel at
31 °C; V: C16TAB in 0.03 M Fe(NO3)3 aqueous media with electrode of copper at 32 °C. (After Refs. [94,186]).
Table 4
Experimental conditions for different surfactant testing systems. cmc and sac are estimated values based on experiment [94,186].
Testing system Surfactant mixed ratio Salt C (M) T (°C) pH Electrode Rotation (RPM) cmc (lM) sac (lM)
I C14BzCl/C14BzCl/C16BzCl NaCl 40 4 X65 steel 300 140 72
0.70/0.25/0.05 0.171
II C14BzCl/C14BzCl/C16BzCl NaCl 40 5 X65 steel 100 16.5 9
0.33/0.33/0.33 0.599
III AAOA NaCl 25 6 1018 steel Low 15 8.2
1 0.856
IV CPC HCl 31 0 1018 steel 1000 1.5 1
1 1
V C16TABr Fe(NO3)3 32 – Copper 1000 30 20
1 0.03
AAOA: N-[2-[(2-aminoethyl) amino] ethyl]-9-octadecenamide; CPC: cetylpyridinium chloride; C16TABr: Hexadecyl trimethyl ammonium bromide.
It has been reported recently that a multiphysics perspective for the performance of surfactant inhibitors in WOS envi-
ronment is proposed following the systematic domains and processes approach, in which, according to the authors, several
physical and chemical processes contributing to the ultimate surfactant inhibitor efficiency are evaluated [104]. However,
data reported for the validation of such a model is very limited, and a tremendous amount of experimental and modeling
work remains to be implemented before ultimate application.
9.3.2. ICI model

As described in previous sections, the inhibition performance of surfactants is usually affected by different phenomena
(see Fig. 22) such as surfactant partitioning between oil and water, micellization, surfactant-corrosion product precipitation,
adsorption/desorption, fluid flow, surfactant-solvent interactions, surfactant-counterion pairing, and lateral interactions
between surfactant molecules. These phenomena are incorporated into three major processes: partitioning between oil
and water, micellization/precipitation, and effective adsorption on metal surfaces. Fluid flow can be simulated using a rotat-
ing disc test. The effects of these factors on corrosion can be modeled.
When the surfactant concentration reaches the cmc, bilayers/multilayers/micelles adsorb on metal surface and micelles
also form in water phase or oil phase or both phases [95]. Steel corrosion inhibition is directly determined by the monomeric
surfactant adsorption and effective surface coverage on metal. The adsorption process, however, is affected by the other two
major processes: partitioning and micellization, which tend to deplete monomeric surfactant availability for adsorption and
effective coverage on metal [181]. Therefore, it is necessary to consider these three major processes simultaneously for a sys-
tematic evaluation and modeling of steel corrosion inhibition using mixed surfactants.
With these in mind, an integrated corrosion inhibition (ICI) model [181], has been provided very recently for the modeling
and prediction of corrosion inhibition efficiency of mixed surfactant inhibitors based on the integration of a few existing
models, including the water-oil surfactant distribution model [95], the aqueous cmc prediction model [89,91,94,136,250],
and the MLA model (or the MQSAR model) [89,91,94]. The developed ICI model has been validated using existing experimen-
tal data and literature reported results, which demonstrate the robustness of this method for integrated corrosion inhibition
prediction.
At the partitioning equilibrium, all of the mixed surfactants in WOS environments should conform to a mass balance
equation [181]
Mtol ¼ Mw þ M o þ M ad ð119Þ
where Mtol is the total quantity of all surfactants added to the WOS environments, Mw is the surfactants (both monomeric
and micellar forms) distributed in the water phase, Mo is the surfactants (both monomeric and micellar forms) distributed in
the oil phase, and Mad is the surfactants adsorbed on metal surface and water/oil interface. The amount of surfactants
adsorbed on the water/oil interface is unlikely to significantly impact the mass balance [236] and thus has become neglected
in Eq. (119). Similarly, the amount of adsorbed surfactants in the form of monolayer, bilayers, multilayers, and micelles on
steel surface is also negligible. Therefore, Eq. (1) can be simplified to the following format without compromise of surfactant
mass balance in WOS environments [181]:
Mtol ¼ Mw þ M o ð120Þ
Alternatively, Eq. (120) is equivalent to Eq. (105) in the water-oil surfactant distribution model as described earlier.
Therefore, the surfactant distribution in WOS environments (three phases) reduces to a distribution phenomenon in
water-oil environments (two phase) which can be evaluated by water-oil surfactant distribution model as introduced in Sec-
tion 7.3. In this regard, the apparent cmc of mixed surfactants in WOS environments (Cap ), monomer concentration of sur-
factant ‘i’ in water and in oil phases (C w o
mi and C mi ), the apparent partitioning coefficient of the mixture (Kmix), and the molar
fraction of surfactant ‘i’ in mixed micelles (ai), can be predicted at given inputs, including partitioning coefficient Ki, the
aqueous cmc Cw i of pure surfactant ‘i’, and the known Ctol and xi, which is the molar fraction of surfactant ‘i’ in total mixed
surfactants in bulk solution [181]. Ki can be determined by the partitioning coefficient determination method which is devel-
oped and validated in previous work [95]. The aqueous cmc of pure and mixed surfactants in salt solution can be calculated
using the improved traditional cmc model [89,91] or molecular thermodynamic cmc model (AMT model) [94,136,250].
Based on the described ICI model, a schematic flow chart is presented in Scheme 1 to illustrate how corrosion inhibition
efficiency g (%) of mixed surfactants in WOS environments is predicted [9,181]. First, mixed surfactants are added to the
aqueous phase in corrosive environments of WOS. At the equilibrium of surfactant partitioning, aggregation, and adsorption,
the surfactant distribution model combined with the cmc model is then used to calculate the equilibrium monomeric con-
centrations of each mixed surfactant component ‘i’ in water and oil phases. The MLA model and the cmc model along with
the calculated results from previous steps are used to evaluate surfactant adsorption/coverage on the metal surface. Finally g
(%) is predicted using Eq. (121) or Eq. (122) with the assumption of the effective surface coverage (mainly due to monolayer)
is equal to corrosion inhibition [9,181].
!
1
g ð%Þ ¼ 100 h ¼ 1 w 100 ð121Þ
1 þ K 0 CCmw
Scheme 1. Flow chart for the use of ICI model to predict corrosion inhibition efficiency of mixed surfactants in WOS environments (After Refs. [9,181]).
! !
ðA0 Q þB0 ÞC w
g ð%Þ ¼ ! !
m
100 ð122Þ
Cw þ ðA0 Q þB0 ÞC w
m
The validation of the developed ICI model and associated comparison to experimental data which are obtained from dif-
ferent testing systems as listed in Table 5 are introduced below [181]. It is found that all the predicted aqueous cmc values of
various pure surfactants and associated mixtures match experimental data very well in Fig. 38(a) and (b) [181]. Note that for
each surfactant mixture in Fig. 38(b), the aqueous cmc varied before (solid symbol) and after (open symbol) partitioning due
to the change in bulk mixed molar ratio of surfactants, which is originally caused by the selective partitioning of a certain
amount of surfactants into oil phase. The new aqueous cmc Cw after partitioning is constant if no micelle formation occurs in
WOS environments (C < Cap or Ctol 6 1:5Cap in the case of Vo/Vw = 1/2). The new Cw would vary as a function of Ctol if
Ctol > 1:5Cap where the bulk mixed molar ratio is affected by both partitioning and micellization. Eventually, the change
in the micellar mixed molar ratio as a function of Ctol leads to the change in the equilibrium bulk mixed molar ratio and thus
the change in Cw.
An excellent agreement is observed between the predicted and the experimental values of Ki as shown in Fig. 38(c) [181].
Fig. 39(a) presents the corrosion current density as a function of time for the surfactant mixture (Ctol = 1.50 105 M) in
0
Testing System (TS) 1 [181]. The concentration of Ctol = 1.50 105 M which is slightly less than Cw and a lot less than
1:5C assures the absence of micelles in the tested environments (both aqueous phase without oil and WOS environments)
ap
and thus facilitates the applicability of ICI model to the discussed testing system. The predicted current density for this test-
ing system agrees with the experimental data reasonably well. The slight underestimation in the presence of oil is due to the
slight overestimation of monomeric surfactants present in bulk aqueous phase because of the neglect of corrosion product-
surfactant precipitates and adsorbed surfactants at the water/oil interface. The overall prediction of corrosion inhibition effi-
Table 5
Testing systems of mixed surfactant partitioning in water-oil environment and associated inhibition on steel corrosion in water. (After Ref. [181]).
Testing system (TS) Surfactants Mixed molar ratio in water NaCl, M T, °C Oil Volume ratio (oil/water)
Initial ratio After partitioninga
1 C12/C14/C16 1/1/1 20/7/1 0.599 40 Toluene 0/1, 1/2, 2/1
2 C12/C14/C16 6/3/1 35/14/1 0.0855 40 Toluene 1/2
3 C16/C16E20 1/1 0.36/1 0.03 25 Heptane 1/2
4 C16TABr/C16/C16E20 1/3/6 0.14/0.33/1 0.03 25 Heptane 1/2
a
Note that the mixed molar ratio in the bulk after partitioning is constant if no micelle formation. Once micelle forms (if Ctol is high enough), the ratio
changes as a function of Ctol. Here it is limited to the former case.
Fig. 38. Comparison of predicted aqueous cmc and experimental aqueous cmc of (a) various pure surfactants and (b) different testing systems of mixed
surfactants NaCl-containing aqueous solution. (c) Comparison of predicted and experimental partitioning coefficients of pure surfactants. The number in
the legend next to the symbols in (a) and (c) is NaCl concentration in unit M. Solid symbols in (b) represent the initial aqueous cmc of surfactant mixture
0
before partitioning, Cw ; open symbols in (b) represent the equilibrated aqueous cmc of surfactant mixture at partitioning and aggregation equilibrium, Cw.
Note for mixtures in (b), Ctol < 1:5Cap , which indicates no micelle formation in WOS environments. 1.5 is calculated using mass balance and oil/water
volume ratio (1:2). (After Ref. [181]).
Fig. 39. (a) Comparison between the predicted and the experimental corrosion current densities of steel samples as a function of exposure time in aqueous
phase in WOS environment of Testing System 1 in Table 5. (b) Comparison of the ICI model predicted and the experimental inhibition efficiency on steel
corrosion of different mixed surfactant systems in WOS environments (water/oil volume ratio is 2) in Table 5. Symbols: experiment; curves: prediction. TS
represents Testing System. Vertical lines in (b) represent Ctol = 1:5Cap , indicating micelle formation concentration in WOS environment. K0 = 4.84, 13.74, and
23.59 for C16TABr, BAC, and C16E20, respectively. The molar fraction-based average value of K0 is used for each TS. (After Ref. [181]).
ciency as a function of Ctol matches experimental data well for all testing systems, as shown in Fig. 39(b) [181]. The slightly
higher prediction of g ð%Þ at Ctol < 1:5Cap can be explained by the same reason for the underestimation of corrosion current
density mentioned previously. On the other hand, g ð%Þ is slightly underestimated at Ctol > 1:5Cap due to micelle formation in
WOS environments where the ICI model is no longer applicable and the predicted g ð%Þ becomes constant as Ctol increases. In
reality g ð%Þ might increase with increasing Ctol due to potential extra metal surface coverage by multilayers/micelles. How-
ever, g ð%Þ is usually high enough (provided the discussed surfactants act as good inhibitors) around Ctol = 1:5Cap , where a
monolayer covers the metal surface well and multilayers/micelles coverage does not contribute much to additional corrosion
inhibition [9]. Therefore, experimental data is only slightly higher than the prediction.
10. Conclusions and perspectives
Corrosion of iron and steel pipes that transport oil and aqueous solutions is a significant industrial challenge. Corrosion in
steel inside pipelines is often driven by the hydrogen ion availability because the more powerful oxidant, oxygen, is less
available. The steel surface often contains a thin barrier layer of iron oxide and an outer layer of iron oxide/hydroxide. How-
ever, if the pH is sufficiently low, bare metal can be exposed to solution. Dissolved compounds/ions such as H2S and Cl in
corrosive media promote the damage of metals. A reduction in the corrosion rate can be achieved by limiting the access of
the oxidant to the steel surface. Surfactants have a high affinity for adsorbing on surfaces and can provide a barrier to further
corrosion.
Corrosion inhibition using surfactants has been studied extensively. A large quantity of research literature describes the
corrosion inhibition performance of a wide array of surfactants. Surfactant adsorption is driven by concentration and amphi-
philic properties of surfactant as well as surface properties such as charge, defects, and composition. The hydrophilic-
hydrophobic nature of surfactants results in interfacial adsorption and aggregation. Surfactants tend to form sub-
monolayer, monolayer/hemi-micelle, or bilayer/cylindrical micelle structures at interfaces and surfaces. Corrosion inhibition
increases only modestly above monolayer level coverage. The concentration needed to achieve monolayer coverage is usu-
ally close to the cmc for the surfactant. Aggregation of surfactant above the cmc leads to the formation of spherical micelles
and other structures. These structures act like a buffer that maintains the free monomeric surfactant molecule concentration
constant when the total surfactant concentration exceeds the level needed for micelle formation. The free monomeric sur-
factant concentration determines the level of adsorption on a surface. Consequently, surfactant induced corrosion inhibition
increases only modestly above the cmc. It is possible to predict corrosion inhibition resulting from surfactant adsorption
based on material properties, surfactant properties such as hydrocarbon chain length, and solution conditions such as pH
and ionic strength.
Various models, including semi-empirical models, mechanistic models, combined semi-empirical and mechanistic mod-
els, and multiphysics models, have been developed for the evaluation of corrosion inhibition using surfactant inhibitors in
aqueous phase and/or in WOS environments. Such a model is the multiphysics ICI model, which is intended to serve as a
basic framework in the understanding of mixed surfactant inhibitor performance with a focus on the application in salt-
containing WOS environments. Beyond this, other potential applications may be extended to the design of surfactants, selec-
tion of optimal surfactants for specific applications, experimental validation of developed models, simulation of relevant pro-
cesses and phenomena, and the integration into more comprehensive lifetime prediction models in which factors affecting
surfactant-based corrosion inhibition efficiency may be evaluated.
Fundamental analyses using relatively new tools such as molecular and quantum chemical calculation also provide
insights into corrosion inhibition. Although these techniques have not provided quantitative inhibition predictions at desired
levels, they have significant potential to achieve better predictions as computing tools improve.
There are several challenges to more accurate fundamental modeling and predictions of corrosion inhibition: (1) The sur-
face state of corroding metals such as steels is dynamic and can include different phases as well as metal inclusions and
alloying elements that are difficult to incorporate into fundamental modeling. (2) The mechanism of ion/counterion binding
to micelle surface and liquid-metal interface is not clear and has been a controversial issue. The ion effects on surface wetting
and thus on surface adsorption of surfactants needs more study. More complex molecular or multivalent ions, such as diva-
lent ions can in principle be modeled as well, but experimental verification and model development are needed. (3) Exten-
sive thermodynamic modeling work on surfactant aggregation and adsorption has been performed, but the research on
surfactant adsorption kinetics and surfactant diffusion is limited. Adsorption kinetics and the mass transfer coefficients
can be studied through the measurement of equilibrium surface tension and surface tension relaxation profiles at different
surfactant concentrations. (4) More experimental work at elevated temperature and higher pressure to more closely simu-
late field conditions should be performed for additional validation of thermodynamic modeling on surfactant aggregation,
adsorption, partitioning, and additional surfactant-associated processes. More realistically, a flow loop should be setup to
simulate field conditions, including flow conditions, temperature, pressure, salinity, pH, mixed ions, shear stress, and pipe
length scale, etc., to provide more reliable data for modeling.
Acknowledgements
The presented research is supported by a grant from BP under contract CW1901503. CHPC at the University of Utah is
acknowledged for providing computational resources. We also gratefully acknowledge discussions with our national/inter-
national colleagues: Dr. Jan D. Miller, Dr. Jim Muller, Dr. Anita Orendt, Dr. Paul D.I. Fletcher, Dr. Alexey I. Victorov, Dr. Daniel
Blankschtein, and Dr. Matjaž Finšgar, and many more.
References
[1] Tribollet B, Kittel J, Meroufel A, Ropital F, Grosjean F, Sutter EMM. Corrosion mechanisms in aqueous solutions containing dissolved H2S. Part 2: model
of the cathodic reactions on a 316L stainless steel rotating disc electrode. Electrochim Acta 2014;124:46–51.
[2] Zou S, Li X, Dong C, Ding K, Xiao K. Electrochemical migration, whisker formation, and corrosion behavior of printed circuit board under wet H2S
environment. Electrochim Acta 2013;114:363–71.
[3] Bai P, Zhao H, Zheng S, Chen C. Initiation and developmental stages of steel corrosion in wet H2S environments. Corros Sci 2015;93:109–19.
[4] Zhou Y, Zuo Y. The intergranular corrosion of mild steel in CO2 + NaNO2 solution. Electrochim Acta 2015;154:157–65.
[5] López DA, Simison SN, de Sánchez SR. The influence of steel microstructure on CO2 corrosion. EIS studies on the inhibition efficiency of benzimidazole.
Electrochim Acta 2003;48:845–54.
[6] George KG, Nešić S. Investigation of carbon dioxide corrosion of mild steel in the presence of acetic acid-part 1: basic mechanisms. Corrosion
2007;63:178–87.
[7] Linter BR, Burstein GT. Reaction of pipeline steel in carbon dioxide solutions. Corros Sci 1999;41:117–39.
[8] Thompson NG, Mark Y, Daniel D. Cost of corrosion and corrosion maintenance strategies. Corros Rev 2007;25:247–62.
[9] Zhu Y. Integrated modeling of mixed surfactants distribution and corrosion inhibition performance in oil pipelines PhD dissertation. University of
Utah; 2015.
[10] Robie RA, Hemingway BS, Fisher JR. Thermodynamic properties of minerals and related substances at 298.15 K and 1 bar (105 Pascals) pressure and at
higher temperatures. Washington, D.C.: Geological Survey Bulletin 1452, U.S. Department of the Interior; 1978.
[11] Garrels RM, Christ CL. Solutions, minerals, and equilibria. Boston: Jones and Bartlett Publishers; 1990.
[12] Abbott AP, Ahmed EI, Harris RC, Ryder KS. Evaluating water miscible deep eutectic solvents (DESs) and ionic liquids as potential lubricants. Green
Chem 2014;16:4156–61.
[13] Gabler C, Tomastik C, Brenner J, Pisarova L, Doerr N, Allmaier G. Corrosion properties of ammonium based ionic liquids evaluated by SEM-EDX, XPS
and ICP-OES. Green Chem 2011;13:2869–77.
[14] Liu J, Macdonald DD. The passivity of iron in the presence of ethylenediaminetetraacetic acid: II. The defect and electronic structures of the barrier
layer. J Electrochem Soc 2001;148:B425–30.
[15] Jolivet JP, Chaneae C, Tronc E. Iron oxide chemistry. From molecular clusters to extended solid networks. Chem Commun 2004:481–3.
[16] Cabrera-Sierra R, Pech-Canul MA, Gonzalez I. The role of hydroxide in the electrochemical impedance response of passive films in corrosion
environments. J Electrochem Soc 2006;153:B101–7.
[17] Monnier J, Réguer S, Foy E, Testemale D, Mirambet F, Saheb M, et al. XAS and XRD in situ characterisation of reduction and reoxidation processes of
iron corrosion products involved in atmospheric corrosion. Corros Sci 2014;78:293–303.
[18] Yamamoto T, Fushimi K, Miura S, Konno H. Influence of substrate dislocation on passivation of pure iron in pH 8.4 borate buffer solution. J
Electrochem Soc 2010;157:C231–7.
[19] Hu J, Koleva DA, de Wit JHW, Kolev H, van Breugel K. Corrosion performance of carbon steel in simulated pore solution in the presence of micelles. J
[20] Cornell RM, Schwertmann U. The iron oxides: structure, properties, reactions, occurrences, and uses. 2nd ed. Darmstadt: Wiley-VCH; 2003.
[21] Nešić S, Postlethwaite J, Olsen S. An electrochemical model for prediction of corrosion of mild steel in aqueous carbon dioxide solutions. Corrosion
1996;52:280–94.
[22] Nešić S, Lunde L. Carbon dioxide corrosion of carbon steel in two-phase flow. Corrosion 1994;50(9):717–27.
[23] Pfennig A, Wiegand R, Wolf M, Bork CP. Corrosion and corrosion fatigue of AISI 420C (X46Cr13) at 60 °C in CO2-saturated artificial geothermal brine.
Corros Sci 2013;68:134–43.
[24] Nešić S. Key issues related to modelling of internal corrosion of oil and gas pipelines – a review. Corros Sci 2007;49:4308–38.
[25] Videm K, Dugstad A. Corrosion of carbon steel in an aqueous carbon dioxide environment-part 2. Mater Perform 1989;4:46–50.
[26] Mu LJ, Zhao WZ. Investigation on carbon dioxide corrosion behavior of HP13Cr110 stainless steel in simulated stratum water. Corros Sci
2010;52:82–9.
[27] Banas J, Lelek-Borkowska U, Mazurkiewicz B, Solarski W. Effect of CO2 and H2S on the composition and stability of passive film on iron alloy in
geothermal water. Electrochim Acta 2007;52:5704–14.
[28] Mishra B, Al-Hassan S, Olson DL, Salama MM. Development of a predictive model for activation-controlled corrosion of steel in solutions containing
carbon dioxide. Corrosion 1997;53:852–9.
[29] Kermani MB, Morshed A. Carbon dioxide corrosion in oil and gas production – a compendium. Corrosion 2003;59:659–83.
[30] Al-Hassan S, Mishra B, Olson DL, Salama MM. Effect of microstructure on corrosion of steels in aqueous solutions containing carbon dioxide. Corrosion
1998;54:480–91.
[31] Medina Huerta JM, Godínez JG, González JL. Corrosion rate prediction in the liquid collection section of separating tanks—part 1: dependence on
temperature, O2, pH, and superficial conditions. Corrosion 2012;68:1–12.
[32] Waard C, Milliams DE. Carbonic acid corrosion of steel. Corrosion 1975;31:177–81.
[33] Ezuber MH. Influence of temperature and thiosulfate on the corrosion behavior of steel in chloride solutions saturated in CO2. Mater Des
2009;30:3420–7.
[34] Waard C, Lotz U, Dugstad A. Influence of liquid flow velocity on CO2 corrosion a semi-empirical model. In: NACE international corrosion conference;
1995. Paper no. 128.
[35] Cardoso Filho JC, Orazem ME. Application of a submerged impinging jet to investigate the influence of temperature, dissolved CO, and fluid velocity on
corrosion of pipeline-grade steel in brine. In: NACE international corrosion conference; 2001. Paper no. 1058.
[36] Liu QY, Maoa LJ, Zhou SW. Effects of chloride content on CO2 corrosion of carbon steel in simulated oil and gas well environments. Corros Sci
2014;84:165–71.
[37] Chen CF, Lu MX, Sun DB, Zhang ZH, Chang W. Effect of chromium on the pitting resistance of oil tube steel in a carbon dioxide corrosion system.
Corrosion 2005;61:594–601.
[38] Videm K, Koren AM. Corrosion, passivity, and pitting of carbon steel in aqueous solutions of HCO
3 , CO2, and Cl . Corrosion 1993;49:746–54.
[39] Brown B, Parkala SR, Nešić S. CO2 corrosion in the presence of trace amounts of H2S. In: NACE international corrosion conference; 2004. Paper no.
4736.
[40] Han J, Carey JW, Zhang J. Effect of sodium chloride on corrosion of mild steel in CO2-saturated brines. J Appl Electrochem 2011;41:741–9.
[41] Pfennig A, Bäßler R. Effect of CO2 on the stability of steels with 1% and 13% Cr in saline water. Corros Sci 2009;51:931–40.
[42] Crolet JL. Corrosion in oil and gas production. In: Beranger G, Mazille H, editors. Corrosion and anticorrosion. France: Hermes Science; 2002.
[43] Fang H, Brown B, Nešić S. High salt concentration effects on CO2 corrosion and H2S corrosion. In: NACE corrosion conference; 2010. Paper no. 276.
[44] Mao X, Liu X, Revie RW. Pitting corrosion of pipeline steel in dilute bicarbonate solution with chloride ions. Corrosion 1994;50:651–7.
[45] Sun Y, George K, Nešić S. The effect of Cl and acetic acid on localized CO2 corrosion in wet gas flow. In: NACE corrosion conference; 2003. Paper no.
327.
[46] Ingham B, Ko M, Laycock N, Burnell J, Kappen P, Kimpton JA, et al. In situ synchrotron X-ray diffraction study of scale formation during CO2 corrosion
of carbon steel in sodium and magnesium chloride solutions. Corros Sci 2012;56:96–104.
[47] Niu L, Nakada L. Effect of chloride and sulfate ions in simulated boiler water on pitting corrosion behavior of 13Cr steel. Corros Sci 2015;96:171–7.
[48] Burstein GT, Davies DH. The effects of anions on the behaviour of scratched iron electrodes in aqueous solutions. Corros Sci 1980;20:1143–55.
[49] Burstein GT, Davies DH. The electrochemical behavior of scratched iron surfaces in aqueous solutions. J Electrochem Soc 1981;128:33–9.
[50] Abelev E, Ramanarayanan TA, Bernasek SL. Iron corrosion in CO2/brine at low H2S concentrations: an electrochemical and surface science study. J
[51] Cheng XL, Ma HY, Zhang JP, Chen X, Chen SH, Yang HQ. Corrosion of iron in acid solutions with hydrogen sulfide. Corrosion 1998;54:369–76.
[52] Ma H, Cheng X, Li G, Chen S, Quan Z, Zhao S, et al. Influence of hydrogen sulfide on corrosion of iron under different conditions. Corros Sci
2000;42:1669–83.
[53] Shoesmith DW, Taylor P, Bailey MG, Owen DG. The formation of ferrous monosulfide polymorph during the corrosion of iron by aqueous hydrogen
sulfide. J Electrochem Soc 1980;127:1007–15.
[54] Rickard D, Luther III GW. Chemistry of iron sulfides. Chem Rev 2007;2007(107):514–62.
[55] Bockris OJ, Reddy AKN. Modern electrochemistry. 2nd ed. New York: Kluwer Academic/Plenum Publishers; 2000.
[56] Newman J, Thomas-Alyea KE. Electrochemical systems. 3rd ed. Hoboken: John Wiley & Sons; 2004.
[57] Jones DA. Principles and prevention of corrosion. 2nd ed. New York: Macmillan Publishing Company; 1992.
[58] Free ML. Chemical processing and utilization of metals in aqueous media. 2nd ed. Ann Arbor: Xanedu; 2004.
[59] Free ML. Hydrometallurgy: fundamentals and applications. Hoboken: John Wiley & Sons; 2013.
[60] Ito M. Structures of water at electrified interfaces: microscopic understanding of electrode potential in electric double layers on electrode surfaces.
Surf Sci Rep 2008;63:329–89.
[61] Thiel PA, Madey TE. The interaction of water with surfaces: fundamental aspects. Surf Sci Rep 1987;7:211–385.
[62] Henderson MA. The interaction of water with solid surfaces: fundamental aspects revisited. Surf Sci Rep 2002;46:5–308.
[63] Hodgson A, Haq S. Water adsorption and the wetting of metal surfaces. Surf Sci Rep 2009;64:381–451.
[64] Verdaguer A, Sacha GM, Bluhm H, Salmeron M. Structure and chemistry at aqueous interfaces. Chem Rev 2006;106:1478–510.
[65] Bedürftig K, Völkening S, Wang Y, Wintterlin J, Jacobi K, Ertl G. Vibrational and structural properties of OH adsorbed on Pt(1 1 1). J Chem Phys
1999;111:11147–54.
[66] Madey TE, Netzer FP. The adsorption of H2O on Ni(1 1 1); influence of preadsorbed oxygen on azimuthal ordering. Surf Sci 1982;117:549–60.
[67] Nakamura M, Ito M. Monomer and tetramer water clusters adsorbed on Ru(0 0 0 1). Chem Phys Lett 2000;325:293–8.
[68] Israelachvili JN, Pashley RM. Molecular layering of water at surfaces and origin of repulsive hydration forces. Nature 1983;306:249–50.
[69] Israelachvili JN, Wennerström H. Role of hydration and water structure in biological and colloidal interactions. Nature 1996;379:219–25.
[70] Adamson AW, Gast AP. Physical chemistry of surfaces. 6th ed. New York: John Wiley & Sons; 1997.
[71] Birdi KS, editor. Introduction to electrical interfacial phenomena. Boca Raton: Taylor and Francis; 2010.
[72] Kelly EG, Spottiswood DJ. Introduction to mineral processing. Somerset: John Wiley & Sons; 1982.
[73] Badiea AM, Mohana KN. Effect of temperature and fluid velocity on corrosion mechanism of low carbon steel in presence of 2-hydrazino-4,7-
dimethylbenzothiazole in industrial water medium. Corros Sci 2009;51:2231–41.
[74] Jiang XY, Zheng G, Ke W. Effect of flow velocity and entrained sand on inhibition performances of two inhibitors for CO corrosion of N80 steel in 3%
NaCl solution. Corros Sci 2005;47:2636–58.
[75] Bartlett RW. Solution mining: leaching and fluid recovery of materials. Philadelphia: Gordon and Breach; 1992.
[76] Bird RB, Stewart WE, Lightfoot EN. Transport phenomena. New York: John Wiley & Sons; 1960.
[77] Free ML. Fundamentals of electrometallurgy in aqueous media. JOM 2007;May:28–33.
[78] Wadsworth ME. Principles of leaching. In: Sohn HY, Wadsworth ME, editors. Rate processes of extractive metallurgy. New York: Plenum Press; 1979.
[79] Miller JD. Cementation. In: Sohn HY, Wadsworth ME, editors. Rate processes of extractive metallurgy. New York: Plenum Press; 1979.
[80] Fogler HS. Elements of chemical reaction engineering. Englewood Cliffs: Prentice-Hall; 1986.
[81] Bailey JE, Ollis DF. Biochemical engineering fundamentals. 2nd ed. New York: McGraw-Hill Publishing Company; 1986.
[82] Atkins PW. Physical chemistry. 3rd ed. New York: W.H. Freeman and Co.; 1986.
[83] Treybal RE. Mass transfer operations. 3rd ed. New York: McGraw-Hill Publishing Company; 1987.
[84] Fuchs-Godec R. Effects of surfactants and their mixtures on inhibition of the corrosion process of ferritic stainless steel. Electrochim Acta
2009;54:2171–9.
[85] Tang Y, Yao L, Kong C, Yang W, Chen Y. Molecular dynamics simulations of dodecylamine adsorption on iron surfaces in aqueous solution. Corros Sci
2011;53:2046–9.
[86] Kronberg B. Surfactant mixtures. Curr Opin Colloid Interface Sci 1997;2:456–63.
[87] Free ML. A new corrosion inhibition model for surfactants that more closely accounts for actual adsorption than traditional models that assume
physical coverage is proportional to inhibition. Corros Sci 2004;46:3101–13.
[88] Free ML. Understanding the effect of surfactant aggregation on corrosion inhibition of mild steel in acidic medium. Corros Sci 2002;44:2865–70.
[89] Zhu Y, Free ML, Yi G. Electrochemical measurement, modeling, and prediction of corrosion inhibition efficiency of ternary mixtures of homologous
surfactants in salt solution. Corros Sci 2015;98:417–29.
[90] Riggs OL. Theoretical aspects of corrosion inhibitors and inhibition. In: Nathan CC, editor. Corrosion inhibitors. NACE; 1973.
[91] Zhu Y, Free ML. The effects of surfactant concentration, adsorption, aggregation, and solution conditions on steel corrosion inhibition and associated
modeling in aqueous media. Corros Sci 2015;102:233–50.
[92] Kokalj A, Peljhan S, Finsgar M, Milosev I. What determines the inhibition effectiveness of ATA, BTAH, and BTAOH corrosion inhibitors on copper? J Am
Chem Soc 2010;132:16657–68.
[93] Gece G. Drugs: a review of promising novel corrosion inhibitors. Corros Sci 2011;53:3873–98.
[94] Zhu Y, Free ML, Yi G. Experimental investigation and modeling of the performance of pure and mixed surfactant inhibitors: aggregation, adsorption,
and corrosion inhibition on steel pipe in aqueous phase. J Electrochem Soc 2015;168:C582–91.
[95] Zhu Y, Free ML. Experimental investigation and modeling of the performance of pure and mixed surfactant inhibitors: partitioning and distribution in
water-oil environments. J Electrochem Soc 2015;162:C702–17.
[96] Free ML, Wang W, Ryu DY. Prediction of corrosion inhibition using surfactants. Corrosion 2004;60:837–44.
[97] Wang W, Free ML. Prediction and measurement of corrosion inhibition of mild steel using nonionic surfactants in chloride media. Corros Sci
2004;46:2601–11.
[98] Wills BA. Mineral processing technology. 3rd ed. Oxford: Pergamon Press; 1985.
[99] Ryu DY, Free ML. The use of electrochemical noise measurements for determining the rate of corrosion and surfactant aggregation transition
concentration at the mild steel-liquid interface. Adsorpt Sci Technol 2004;22:155–64.
[100] Moises de Oliveira HP, Gehlen MH. Characterization of mixed micelles of sodium dodecyl sulfate and tetraoxyethylene dodecyl ether in aqueous
solution. Langmuir 2002;18:3792–6.
[101] Zhang R, Somasundaran P. Advances in adsorption of surfactants and their mixtures at solid/solution interfaces. Adv Coll Interface Sci 2006;123–
126:213–29.
[102] Migahed MA, Hegazy MA, Al-Sabagh AM. Synergistic inhibition effect between Cu2+ and cationic Gemini surfactant on the corrosion of downhole
tubing steel during secondary oil recovery of old wells. Corros Sci 2012;61:10–8.
[103] Durnie W, De Marco R, Jefferson A, Kinsella B. Development of a structure-activity relationship for oil field corrosion inhibitors. J Electrochem Soc
1999;146:1751–6.
[104] Taylor CD, Chandra A, Vera J, Sridhar N. A multiphysics perspective on mechanistic models for chemical corrosion inhibitor performance. J
[105] Christov M, Popova A. Adsorption characteristics of corrosion inhibitors from corrosion rate measurements. Corros Sci 2004;46:1613–20.
[106] Okafor PC, Zheng Y. Synergistic inhibition behaviour of methylbenzyl quaternary imidazoline derivative and iodide ions on mild steel in H2SO4
solutions. Corros Sci 2009;51:850–9.
[107] Behpour M, Ghoreishi SM, Soltani N, Salavati-Niasari M, Hamadanian M, Gandomi A. Electrochemical and theoretical investigation on the corrosion
inhibition of mild steel by thiosalicylaldehyde derivatives in hydrochloric acid solution. Corros Sci 2008;50:2172–81.
[108] Shi X, Zhang R, Minot C, Hermann K, Van Hove MA, Wang W, et al. Complex molecules on a flat metal surface: large distortions induced by
chemisorption can make physisorption energetically more favorable. J Phys Chem Lett 2010;1:2974–9.
[109] Israelachvili JN. Intermolecular and surface forces. 3rd ed. San Diego: Academic Press; 2011.
[110] Barnes G, Gentle I. Interfacial science: an introduction. Oxford: Oxford University Press; 2005.
[111] Stoyanov SD, Paunov VN, Rehage H, Kuhn H. A new class of interfacial tension isotherms for nonionic surfactants based on local self-consistent mean
field theory; classical isotherms revisite. Phys Chem Chem Phys 2004;6:596–603.
[112] Valkovska DS, Shearman GC, Bain CD, Darton RC, Eastoe J. Adsorption of ionic surfactants at an expanding air-water interface. Langmuir
2004;20:4436–45.
[113] Diez-Pascual AM, Compostizo A, Crespo-Colin A, Rubio RG, Miller R. Adsorption of water-soluble polymers with surfactant character: adsorption
kinetics and equilibrium properties. J Colloid Interface Sci 2007;307:398–404.
[114] Szpakowska M, Magnuszewska A, Nagy OB. Mechanism of nitromethane liquid membrane oscillator containing sodium oleate. J Colloid Interface Sci
2008;325:494–9.
[115] Dretschkow T, Wandlowski T. Adsorption of organic substances on the free water surface and at the water-mercury interface. Surf Sci Rep
2009;64:381–451.
[116] Zhou Q, Rosen MJ. Molecular interactions of surfactants in mixed monolayers at the air/aqueous solution interface and in mixed micelles in aqueous
media; the regular solution approach. Langmuir 2003;19:4555–62.
[117] Atkina R, Craig VSJ, Wanless EJ, Biggs S. Mechanism of cationic surfactant adsorption at the solid–aqueous interface. Adv Coll Interface Sci
2003;103:219–304.
[118] Schram V, Hall SB. Thermodynamic effects of the hydrophobic surfactant proteins on the early adsorption of pulmonary surfactant. Biophys J
2001;81:1536–46.
[119] Free ML, Miller JD. Kinetics of 18-carbon carboxylate adsorption at the fluorite surface. Langmuir 1997;13:4377–82.
[120] Mortimer RG. Physical chemistry. California: The Benjamin/Cummings Publishing Company; 1993.
[121] Butt H, Graf K, Kappl M. Physics and chemistry of interfaces. Weinheim: Wiley-VCH Verlag GmbH & Co. KGaA; 2003.
[122] Hunter RJ. Foundations of colloid science. 2nd ed. New York: Oxford University Press; 2001.
[123] Shinoda K, Hato M, Hayashi T. Physicochemical properties of aqueous solutions of fluorinated surfactants. J Phys Chem 1972;76:909–14.
[124] Ozeki S, Tsunoda M, Ikeda S. Surface tension of aqueous solutions of dodecyl dimethyl ammonium chloride, and its adsorption on aqueous surfaces. J
Colloid Interface Sci 1978;64:28–35.
[125] Goodwin JW. Colloids and interfaces with surfactants and polymers-introduction. England: Jon Wiley & Sons; 2004.
[126] Drelich J. Role of wetting phenomena in the hot water process for bitumen recovery from tar sand PhD dissertation. University of Utah; 1993.
[127] Derjaguin BV, Landau L. Theory of the stability of strongly charged lyophobic sols and the adhesion of strongly charged particles in solutions of
electrolytes. Acta Physicochem URSS 1941;14:633–62.
[128] Shchukin ED, Pertsov AV, Amelina EA, Zelenev AS. Colloid and surface chemistry. The Netherlands: Elsevier; 2001.
[129] Kolasinski KW. Surface science. England: John Wiley & Sons; 2002.
[130] Schramm LL. Emulsions, foams, and suspensions: fundamentals and applications. Wiley-VCH; 2005.
[131] Kodama M, Seki S. Thermodynamical investigations on phase transitions of surfactant-water systems. Adv Coll Interface Sci 1991;35:1–30.
[132] Rusanov AI. Micellization in surfactant solutions. Readings: Harwood Academic Publishers; 1997.
[133] Oh SG, Shah DO. The effect of micellar lifetime on the rate of solubilization and detergency in sodium dodecyl sulfate solutions. JAOCS
1993;70:673–8.
[134] Tanford C. The hydrophobic effect: formation of micelles and biological membranes. 2nd ed. New York: Wiley Interscience; 1980.
[135] Koroleva SV, Victorov AI. Modeling of the effects of ion specificity on the onset and growth of ionic micelles in a solution of simple salts. Langmuir
2014;30:3387–96.
[136] Zhu Y, Free ML. Experimental investigation and modeling of the performance of pure and mixed surfactant inhibitors: micellization and corrosion
inhibition. Colloids Surf A: Physicochem Eng Aspects 2016;489:407–22.
[137] Jonsson B, Lindman B, Holmberg K, Kronberg B. Surfactants and polymers in aqueous solution. Chichester: Wiley; 1998.
[138] Shinoda K, Becher P. Principles of solution and solubility. New York: Marcel Dekker; 1978.
[139] Chhabra V, Free ML, Kang PK, Truesdail SE, Shah DO. Microemulsions as an emerging technology: from petroleum recovery to nanoparticle synthesis
of magnetic materials and superconductors. Tenside Surfactants Deterg 1997;34:156–68.
[140] Nagarajan R, Ruckenstein E. Molecular theory of microemulsions. Langmuir 2000;16:6400–15.
[141] Finšgar M, Jackson J. Application of corrosion inhibitors for steels in acidic media for the oil and gas industry: a review. Corros Sci 2014;86:17–41.
[142] Malik MA, Hashim MA, Nabi F, Al-Thabaiti SA. Anti-corrosion ability of surfactants: a review. Int J Electrochem Sci 2011;6:1927–48.
[143] Williams DA, Holifield PK, Looney JR, McDougall LA. US patent: 5,200,096. Exxon Chemicals Patents, Inc; 1993.
[144] Ali S, Reyes JS, Samuel MM, Auzerais FM. US patent: 2010/0056405 A1. Schlumberger Technology Corporation; 2010.
[145] Nasr-El-Din HA, Al-Othman AM, Taylor KC, Al-Ghamdi AH. Surface tension of HCl-based stimulation fluids at high temperatures. J Petrol Sci Eng
2004;43:57–73.
[146] Hill DG, Romijn H. Reduction of risk to the marine environment from oilfield chemicals: environmentally improved acid corrosion inhibition for well
stimulation. In: NACE corrosion conference; 2000. Paper no. 00342.
[147] Horsup DI, Clark JC, Binks BP, Fletcher PDI, Hicks JT. The fate of oilfield corrosion inhibitors in multiphase systems. Corrosion 2010;66:036001–14.
[148] Hill RM. Mixed surfactant systems. New York: Marcel Dekker; 1993.
[149] Rosen MJ. Surfactants and interfacial phenomena. 3rd ed. New Jersey: John Wiley & Sons; 2004.
[150] Heydari M, Javidi M. Corrosion inhibition and adsorption behaviour of an amido imidazoline derivative on API 5L X52 steel in CO2-saturated solution
and synergistic effect of iodide ions. Corros Sci 2012;61:148–55.
[151] Okafor PC, Liu CB, Liu X, Zheng YG. Inhibition of CO2 corrosion of N80 carbon steel by carboxylic quaternary imidazoline and halide ions additives. J
Appl Electrochem 2009;39:2535–43.
[152] Oguzie EE, Li Y, Wang FH. Corrosion inhibition and adsorption behavior of methionine on mild steel in sulfuric acid and synergistic effect of iodide
ion. Corros Sci 2007;310:90–8.
[153] Ebenso EE, Alemu H, Umoren SA, Obot IB. Inhibition of mild steel corrosion in sulphuric acid using alizarin yellow GG dye and synergistic iodide
additive. Int J Electrochem Sci 2008;3:1325–39.
[154] Bouklah M, Hammouti B, Aouniti A, Benkaddour M, Bouyanzer A. Synergistic effect of iodide ions on the corrosion inhibition of steel in 0.5 M H2SO4
by new chalcone derivatives. Appl Surf Sci 2006;252:6236–42.
[155] Larabi L, Harek Y, Traisnel M, Mansri A. Synergistic influence of poly(4-vinylpyridine) and potassium iodide on inhibition of corrosion of mild steel in
1 M HCl. J Appl Electrochem 2004;34:833–9.
[156] Shiao SY, Patist A, Free ML, Chhabra V, Huibers PDT, Gregory A, et al. The Importance of sub-angstrom distances in mixed surfactant systems for
technological processes. Colloids Surf A: Physicochem Eng Aspects 1997;128:197–208.
[157] Chhabra V, Free ML, Kang PK, Truesdail SE, Shah DO. Microemulsions as an emerging technology: from petroleum recovery to nanoparticle synthesis
of magnetic materials and superconductors. In: World surfactants congress, 4th, Barcelona; 1996.
[158] Shiao SY, Chhabra V, Patist A, Free ML, Huibers PDT, Gregory A, et al. Chain length compatibility effects in mixed surfactant systems for technological
applications. Adv Coll Interface Sci 1998;74:1–29.
[159] Gao J, Weng Y, Salitanate S, Feng L, Yue H. Corrosion inhibition of a bunsaturated carbonyl compounds on steel in acid medium. Petrol Sci
2009;6:201–7.
[160] Sastri VS. Corrosion inhibitors: principles and applications. Chichester: John Wiley & Sons; 2001.
[161] Baddini ALdQ, Cardoso SP, Hollauer E, Gomes JAdCP. Statistical analysis of a corrosion inhibitor family on three steel surfaces (duplex, super-13 and
carbon) in hydrochloric acid solutions. Electrochim Acta 2007;53:434–46.
[162] Flores EA, Olivares O, Likhanova NV, Domínguez-Aguilar MA, Nava N, Guzman Lucero D, et al. Sodium phthalamates as corrosion inhibitors for carbon
steel in aqueous hydrochloric acid solution. Corros Sci 2011;53:3899–913.
[163] Aljourani J, Raeissi K, Golozar MA. Benzimidazole and its derivatives as corrosion inhibitors for mild steel in 1 M HCl solution. Corros Sci
2009;51:1836–43.
[164] Khaled KF, Babić-Samardžija K, Hackerman N. Piperidines as corrosion inhibitors for iron in hydrochloric acid. J Appl Electrochem 2004;34:697–704.
[165] Tadmouri R, Zedde C, Routaboul C, Micheau JC, Pimienta V. Partition and water/oil adsorption of some surfactants. J Phys Chem B
2008;112:12318–25.
[166] Foo KY, Hameed BH. Insights into the modeling of adsorption isotherm systems. Chem Eng J 2010;156:2–10.
[167] Fuchs-Godec R. The adsorption, cmc determination and corrosion inhibition of some N-alkyl quaternary ammonium salts on carbon steel surface in 2
M H2SO4. Colloids Surf A: Physicochem Eng Aspects 2006;280:130–9.
[168] Farsak M, Keles H, Keles M. A new corrosion inhibitor for protection of low carbon steel in HCl solution. Corros Sci 2015;98:223–32.
[169] Kern MP, Landolt D. Electrochemical and quantum chemical studies of new thiadiazole derivatives adsorption on mild steel in normal hydrochloric
acid medium. Corros Sci 2005;47:485–505.
[170] Obot IB, Macdonald DD, Gasem ZM. Density functional theory (DFT) as a powerful tool for designing new organic corrosion inhibitors. Part 1: an
overview. Corros Sci 2015;99:1–30.
[171] Gece G. The use of quantum chemical methods in corrosion inhibitor studies. Corros Sci 2008;50:2981–92.
[172] Sonu, Tiwari AK, Saha SK. Study on mixed micelles of cationic gemini surfactants having hydroxyl groups in the spacers with conventional cationic
surfactants: effects of spacer group and hydrocarbon tail length. Ind Eng Chem Res 2013;52:5895–905.
[173] Zhao H, Zhang X, Ji L, Hu H, Li Q. Quantitative structure-activity relationship model for amino acids as corrosion inhibitors based on the support
vector machine and molecular design. Corros Sci 2014;83:261–71.
[174] Winkler DA, Breedon M, Hughes AE, Burden FR, Barnard AS, Harvey TG, et al. Towards chromate-free corrosion inhibitors: structure–property models
for organic alternatives. Green Chem 2014;16:3349–57.
[175] Li L, Zhang X, Gong S, Zhao H, Bai Y, Li Q, et al. The discussion of descriptors for the QSAR model and molecular dynamics simulation of benzimidazole
derivatives as corrosion inhibitors. Corros Sci 2015;99:76–88.
[176] Mondal SK, Taylor SR. The identification and characterization of organic corrosion inhibitors: correlation of a computational model with experimental
results. J Electrochem Soc 2014;161:C476–85.
[177] Fischer ER, Parker JE. Technical note: tall oil fatty acid anhydrides as corrosion inhibitor intermediates. Corrosion 1997;53:62–4.
[178] Jovancicevic V, Ahn YS, Dougherty JA, Alink BA. In: NACE corrosion conference 2000; Paper no. 7.
[179] Wang L, Chen W. Density functional theory for adsorption of HCHO on the FeO (1 0 0) surface. J Nat Gas Chem 2010;19:21–4.
[180] Liu H, Teng B, Fan M, Wang B, Zhang Y, Harris HG. CH4 dissociation on the perfect and defective MgO (0 0 1) supported Ni4. Fuel 2014;123:285–92.
[181] Zhu Y, Free ML, Cho J. Integrated evaluation of mixed surfactants’ distribution in water-oil-steel pipe environments and associated corrosion
inhibition efficiency. Corros Sci 2016;110:213–27.
[182] Ryu DY, Free ML. The use of electrochemical quartz crystal microbalance and surface tension measurements for the determination of octylamine and
cetyl pyridinium chloride adsorption in sodium chloride solutions. Colloids Surf A: Physicochem Eng Aspects 2003;226:17–23.
[183] Free ML. The development and application of useful equations to predict corrosion inhibition by different surfactants in various aqueous
environments. Corrosion 2002;58:1025–30.
[184] Zhu Y, Free ML. Mixed-surfactant aggregation, adsorption, and associated steel corrosion inhibition in salt solution. Int J Corros Scale Inhib
2015;4:311–27.
[185] Ning Z, Zhu Y, Free ML. Experimental and modeling investigation of pure and mixed surfactant aggregation and associated steel corrosion inhibition
in aqueous media. Int J Electrochem Sci 2015;10:10462–77.
[186] Zhu Y, Free ML. Electrochemical measurement and modeling of corrosion inhibition efficiency of surfactants in salt solutions applications, monitoring
and models in corrosion. ECS Trans 2015;66:53–72.
[187] Ryu DY, Free ML. The importance of temperature and viscosity effects for surfactant adsorption measurements made using the electrochemical
quartz crystal microbalance. J Colloid Interface Sci 2003;264:402–6.
[188] Wang W, Free ML. Prediction and measurement of mild steel inhibition by alkyl pyridinium and alkyl trimethyl ammonium bromide surfactants in
acidic chloride media. Anti Corros Meth Mater 2003;50:186–92.
[189] Wang W, Free ML, Horsup D. Prediction and measurement of corrosion inhibition of mild steel by imidazoline in brine solutions. Metal Mater Trans B
2005;36:335–41.
[190] Fu JJ, Li SN, Wang Y, Liu XD, Lu LD. Computational and electrochemical studies on the inhibition of corrosion of mild steel by L-cysteine and its
derivatives. J Mater Sci 2011;46:3550–9.
[191] Bilkova K, Gulbrandsena E. Kinetic and mechanistic study of CO2 corrosion inhibition by cetyl trimethyl ammonium bromide. Electrochim Acta
2008;53:5423–33.
[192] Durnie W, Marco RD, Kinsella B, Jefferson A, Pejcica B. Predicting the adsorption properties of carbon dioxide, corrosion inhibitors using a structure-
activity relationship. J Electrochem Soc 2005;152:B1–B11.
[193] Zhang GA, Cheng YF. On the fundamentals of electrochemical corrosion of X65 steel in CO2-containing formation water in the presence of acetic acid
in petroleum production. Corros Sci 2009;51:87–94.
[194] Manne S, Cleveland JP, Gaub HE, Stucky GD, Hansma PK. Direct visualization of surfactant hemimicelles by force microscopy of the electrical double
layer. Langmuir 1994;10:4409–13.
[195] Kunitsugu A, Shimura T. An ultrathin polymer coating of carboxylate self-assembled monolayer adsorbed on passivated iron to prevent iron corrosion
in 0.1 M Na2SO4. Corros Sci 2010;52:1–6.
[196] Migahed M, Nassar MIF. Corrosion inhibition of tubing steel during acidization of oil and gas wells. Electrochim Acta 2008;53:2877–82.
[197] Asefi D, Mahmoodi NM, Arami M. Effect of nonionic co-surfactants on corrosion inhibition effect of cationic gemini surfactant. Colloids Surf A:
Physicochem Eng Aspects 2010;355:183–6.
[198] Duda Y, Govea-Rueda R, Galicia M, Beltrán HI, Zamudio-Rivera LS. Corrosion inhibitors: design, performance, and computer simulations. J Phys Chem
B 2005;109:22674–84.
[199] Singh AK, Shukla SK, Quraishi MA. Inhibitive effect of ceftazidime on corrosion of mild steel in hydrochloric acid solution. Mater Chem Phys
2011;129:68–76.
[200] Olasunkanmi LO, Obot IM, Kabanda MM, Ebenso EE. Some quinoxalin-6-yl derivatives as corrosion inhibitors for mild steel in hydrochloric acid:
experimental and theoretical studies. J Phys Chem C 2015;119:16004–19.
[201] Oguzie EE, Adindu CB, Enenebeaku CK, Ogukwe CE, Chidiebere MA, Oguzie KL. Natural products for materials protection: mechanism of corrosion
inhibition of mild steel by acid extracts of piper guineense. J Phys Chem C 2012;116:13603–15.
[202] Caipa Campos MA, Trilling AK, Yang M, Giesbers M, Beekwilder J, Paulusse JMJ, et al. Self-assembled functional organic monolayers on oxide-free
copper. Langmuir 2011;27:8126–33.
[203] Schwind M, Hosseinpour S, Johnson CM, Langhammer C, Zorić I, Leygraf C, et al. Combined in situ quartz crystal microbalance with dissipation
monitoring, indirect nanoplasmonic sensing, and vibrational sum frequency spectroscopic monitoring of alkanethiol-protected copper corrosion.
Langmuir 2013;29:7151–61.
[204] Kokalj A, Peljhan S. Density functional theory study of ATA, BTAH, and BTAOH as copper corrosion inhibitors: adsorption onto Cu(1 1 1) from gas
phase. Langmuir 2010;26:14582–93.
[205] Smith FN, Taylor CD, Um W, Kruger AA. Technetium incorporation into goethite (a-FeOOH): an atomic-scale investigation. Environ Sci Technol
2015;49:13699–707.
[206] Narayanan TSNS, Park S, Lee MH. Strategies to improve the corrosion resistance of microarc oxidation (MAO) coated magnesium alloys for degradable
implants: prospects and challenges. Prog Mater Sci 2014;60:1–71.
[207] Hanus MJ, Harris AT. Nanotechnology innovations for the construction industry. Prog Mater Sci 2013;58:1056–102.
[208] Hu Y, Li C, Wang X, Yang Y, Zhu H. 1,3,4-Thiadiazole: synthesis, reactions, and applications in medicinal, agricultural, and materials chemistry. Chem
Rev 2014;114:5572–610.
[209] Somasundaran P, Krishnakumar S. Adsorption of surfactants and polymers at the solid-liquid interface. Colloids Surf A: Physicochem Eng Aspects
1997;123–124:491–513.
[210] Nelson N, Schwartz DK. Specific ion (Hofmeister) effects on adsorption, desorption, and diffusion at the solid–aqueous interface. J Phys Chem Lett
2013;4:4064–8.
[211] Bhattacharya S, Samanta SK. Surfactants possessing multiple polar heads. A perspective on their unique aggregation behavior and applications. J Phys
Chem Lett 2011;2:914–20.
[212] Desimone MP, Gordillo G, Simison SN. The effect of temperature and concentration on the corrosion inhibition mechanism of an amphiphilic amido-
amine in CO2 saturated solution. Corros Sci 2011;53:4033–43.
[213] Focke WW, Nhlapo NS, Vuorinen E. Thermal analysis and FTIR studies of volatile corrosion inhibitor model systems. Corros Sci 2013;77:88–96.
[214] Knag M, Sjoblom J, Oye G, Gulbrandsen E. A quartz crystal microbalance study of the adsorption of quaternary ammonium derivates on iron and
cementite. Colloids Surf A: Physicochem Eng Aspects 2004;250:269–78.
[215] Kokalj A. Formation and structure of inhibitive molecular film of imidazole on iron surface. Corros Sci 2013;68:195–203.
[216] Olsen R, Leirvik KN, Kvamme B, Kuznetsova T. Adsorption properties of triethylene glycol on a hydrated {1 0 1 4} calcite surface and its effect on
adsorbed water. Langmuir 2015;31:8606–17.
[217] Khaled KF. Molecular simulation, quantum chemical calculations and electrochemical studies for inhibition of mild steel by triazole. Electrochim Acta
2008;53:3484–92.
[218] Khaled KF. Monte Carlo simulations of corrosion inhibition of mild steel in 0.5 M sulphuric acid by some green corrosion inhibitors. J Solid State
Electrochem 2009;13:1743–56.
[219] Zhang S, Tao Z, Liao S, Wu F. Substitutional adsorption isotherms and corrosion inhibitive properties of some oxadiazol-triazole derivative in acidic
solution. Corros Sci 2010;52:3126–32.
[220] Naderi AHE, Jafari M, Ehteshamzadeh MG, Hosseini G. Effect of carbonsteel microstructures and molecular structure of two new Schiff base
compounds on inhibition performance in 1 M HCl solution by EIS. Mater Chem Phys 2009;115:852–8.
[221] Machnikova E, Whitmire KH, Hackerman N. Corrosion inhibition of carbon steel in hydrochloric acid by furan derivatives. Electrochim Acta
2008;53:6024–32.
[222] Cook A, Gabriel A, Laycock N. On the mechanism of corrosion protection of mild steel with polyaniline. J Electrochem Soc 2004;151:B529–35.
[223] Cruz J, Garcia-Ochoa E, Castro M. Experimental and theoretical study of the 3-amino-1,2,4-triazole and 2-aminothiazole corrosion inhibitors in
carbon steel. J Electrochem Soc 2003;150:B26–35.
[224] Kanojia R, Singh G. An interesting and efficient organic corrosion inhibitor for mild steel in acidic medium. Surf Eng 2005;21:180–6.
[225] Bockris JOM, Devanathan MAV, Muller K. On the structure of charged interfaces. Proc R Soc Lond A 1963;274:55–79.
[226] Hiemenz PC, Rajagopalan R. Principles of colloid and surface chemistry. third ed. New York: Marcel Dekker; 1997.
[227] Aharoni C, Ungarish M. Kinetics of activated chemisorption. Part 2. Theoretical models. J Chem Soc, Faraday Trans 1977;73:456–64.
[228] Haghseresht F, Lu G. Adsorption characteristics of phenolic compounds onto coal-reject-derived adsorbents. Energy Fuels 1998;12:1100–7.
[229] Fuchs-Godec R, Pavlovic MG. Synergistic effect between non-ionic surfactant and halide ions in the forms of inorganic or organic salts for the
corrosion inhibition of stainless-steel X4Cr13 in sulphuric acid. Corros Sci 2012;58:192–201.
[230] Dhar HP, Conway BE, Joshi KM. On the form of adsorption isotherms for substitutional adsorption of molecules of different sizes. Electrochim Acta
1973;18:789–98.
[231] Bockris JOM, Swinkels DAJ. Adsorption of n-decylamine on solid metal electrodes. J Electrochem Soc 1964;111:736–43.
[232] Boer JH. The dynamical character of adsorption. Oxford: Oxford University Press; 1953.
[233] Amin MA, Ahmed MA, Arida HA, Arslan T, Saracoglu M, Kandemirli F. Monitoring corrosion and corrosion control of iron in HCl by non-ionic
surfactants of the TRITON-X series – Part II. Temperature effect, activation energies and thermodynamics of adsorption. Corros Sci 2011;53:540–8.
[234] Do D. Adsorption analysis: equilibria and kinetics. London: Imperial College Press; 1980.
[235] Brown DG, Nuaimi KSA. Nonionic surfactant adsorption onto the bacterial cell surface: a multi-interaction isotherm. Langmuir 2005;21:11368–72.
[236] Kibbey TCG, Chen L. Phase volume effects in the sub- and super-CMC partitioning of nonionic surfactant mixtures between water and immiscible
organic liquids. Colloids Surf A: Physicochem Eng Aspects 2008;326:73–82.
[237] Edwards A, Osborne C, Webster S, Klenerman D, Joseph M, Ostovar P, et al. Mechanistic studies of the corrosion inhibitor oleic imidazoline. Corros Sci
1994;36:315–25.
[238] Ramachandran S, Tsai BL, Blanco M, Chen H, Tang Y, Goddard III WA. In: Jones RH, Bear DR, editors. New techniques for characterizing corrosion and
stress corrosion. The Minerals, Metals and Materials Society; 1996.
[239] Ramachandran S, Tsai B, Blanco M, Chen H, Tang Y, Goddard WA. Self-assembled monolayer mechanism for corrosion inhibition of iron by
imidazolines. Langmuir 1996;12:6419–28.
[240] Hausler RH. On the use of linear polarization measurements for the evaluation of corrosion inhibitors in concentrated HCl at 200 °F (93 °C). Corrosion
1986;42:729–39.
[241] Ulman A. Formation and structure of self-assembled monolayers. Chem Rev 1996;96:1533–54.
[242] Schwartz DK. Mechanisms and kinetics of self-assembled monolayer formation. Annu Rev Phys Chem 2001;52:107–37.
[243] Ramachandran S, Menendez C, Jovancicevic V, Long J. Film persistency of new high-temperature water-based batch corrosion inhibitors for oil and
gas wells. J Petrol Explor Prod Technol 2012;2:125–31.
[244] DeMarco R, Durniev W, Jefferson A, Kinsella B, Crawford A. Persistence of carbon dioxide corrosion inhibitors. Corrosion 2002;58:354–63.
[245] Jang W, Miller JD. Molecular-orientation of Langmuir-Blodgett and self-assembled monolayers of stearate species at a fluorite surface as described by
linear dichroism theory. J Phys Chem 1995;99:10272–9.
[246] Free ML. 18-Carbon unsaturated carboxylate adsorption phenomena at the fluorite surface PhD dissertation. University of Utah; 1994.
[247] Chen S, Frank CW. Infrared and fluorescence spectroscopic studies of self-assembled n-alkanoic acid monolayers. Langmuir 1989;5:978–87.
[248] Free ML, Shah DO. Adsorption and desorption of cetyl pyridinium ions at a tungsten-coated silicon wafer surface. J Colloid Interface Sci
1998;208:104–9.
[249] Zhu Y, Free ML. Mixed-surfactant aggregation and adsorption effects on steel corrosion inhibition in aqueous media with salt. In: Department of
defense-allied nations technical corrosion conference; 2015.
[250] Zhu Y, Free ML. Evaluation of ion effects on surfactant aggregation from improved molecular thermodynamic modeling. Ind Eng Chem Res
2015;54:9052–6.
[251] Phillips JN. The energetics of micelle formation. Trans Faraday Soc 1955;51:561–9.
[252] Bourrel M, Schechter RS. Microemulsions and related systems: formation, solvency, and physical properties. New York: Marcel Dekker; 1988.
[253] Ikeda S, Hayashi S, Imae T. Rodlike micelles um dodecyl sulfate in concentrated sodium halide solutions. J Phys Chem 1981;85:106–12.
[254] Goldsipe A, Blankschtein D. Modeling counterion binding in ionic-nonionic and ionic-zwitterionic binary surfactant mixtures. Langmuir
2005;21:9850–65.
[255] Lima FS, Cuccovia IM, Horinek D, Amaral LQ, Riske KA, Schreiber S, et al. Effect of counterions on the shape, hydration, and degree of order at the
interface of cationic micelles: the triflate case. Langmuir 2013;29:4193–203.
[256] Lima FS, Maximiano FA, Cuccovia IM, Chaimovich H. Surface activity of the triflate ion at the air/water interface and properties of n, n, n-trimethyl-n-
dodecylammonium triflate aqueous solutions. Langmuir 2011;27:4319–23.
[257] Abezgauz L, Kuperkar K, Hassan PA, Ramon O, Bahadur P, Danino D. Effect of Hofmeister anions on micellization and micellar growth of the surfactant
cetylpyridinium chloride. J Colloid Interface Sci 2010;342:83–92.
[258] Oelschlaeger C, Suwita P, Willenbacher N. Effect of counterion binding efficiency on structure and dynamics of wormlike micelles. Langmuir
2010;26:7045–53.
[259] Vlachy N, Jagoda-Cwiklik B, Vácha R, Touraud D, Jungwirth P, Kunz W. Hofmeister series and specific interactions of charged headgroups with
aqueous ions. Adv Coll Interface Sci 2009;146:42–7.
[260] Magid LJ. The surfactant-polyelectrolyte analogy. J Phys Chem B 1998;102:4064–74.
[261] Netz RR, Horinek D. Progress in modeling of ion effects at the vapor/water interface. Annu Rev Phys Chem 2012;63:401–18.
[262] Moreira L, Firoozabadi A. Molecular thermodynamic modeling of specific ion effects on micellization of ionic surfactants. Langmuir
2010;26:15177–91.
[263] Koroleva SV, Victorov AI. The strong specific effect of coions on micellar growth from molecular-thermodynamic theory. Phys Chem Chem Phys
2014;16:17422–5.
[264] Nagarajan R, Ruchenstein E. Theory of surfactant self-assembly: a predictive molecular thermodynamic approach. Langmuir 1991;7:2934–69.
[265] Nagarajan R. In: Esumi K, Ueno M, editors. Theory of micelle formation. Boca Raton: CRC Press; 2003.
[266] Srinivasan V, Blankschtein D. Effect of counterion binding on micellar solution behavior: 1. Molecular-thermodynamic theory of micellization of ionic
surfactants. Langmuir 2003;19:9932–45.
[267] Srinivasan V, Blankschtein D. Effect of counterion binding on micellar solution behavior: 2. Prediction of micellar solution properties of ionic
surfactant-electrolyte systems. Langmuir 2003;19:9946–61.
[268] Srinivasan V, Blankschtein D. Prediction of conformational characteristics and micellar solution properties of fluorocarbon surfactants. Langmuir
2005;21:1647–60.
[269] Andreev VA, Victorov AI. Molecular thermodynamics for micellar branching in solutions of ionic surfactants. Langmuir 2006;22:8298–310.
[270] Manciu M, Ruckenstein E. On the interactions of ions with the air/water interface. Langmuir 2005;21:11312–9.
[271] Lukanov B, Firoozabadi A. Specific ion effects on the self-assembly of ionic surfactants: a molecular thermodynamic theory of micellization with
dispersion forces. Langmuir 2014;30:6373–83.
[272] Zemaitis JF, Clark DM, Rafal M, Scrivner NC. Handbook of aqueous electrolyte thermodynamics. New York: AlChE; 1986.
[273] Pitzer KS. Activity coefficients in electrolyte solutions. 2nd ed. Boca Raton: CRC Press; 1991.
[274] Butler JN. Ionic equilibrium: solubility and pH calculations. New York: John Wiley & Sons; 1998.
[275] Schwarzenbach RP, Gschwend PM, Imboden DM. Environmental organic chemistry. 2nd ed. New York: John Wiley and Sons; 2001.
[276] Jo Nomuranes MN, Piercy J. Light scattering studies on n-dodecyltrimethylammonium bromide and n-dodecylpyridinium iodide. J Chem Soc, Faraday
Trans 1972;1(68):1839–48.
[277] Ozeki S, Ikeda S. The sphere-rod transition of micelles of dodecyldimethylammonium bromide in aqueous NaBr solutions, and the effects of
counterion binding on the micelle size, shape and structure. Colloid Polym Sci 1984;262:409–17.
[278] Imae T, Kamiya R, Ikeda S. Formation of spherical and rod-like micelles of cetyltrimethylammonium bromide in aqueous NaBr solutions. J Colloid
Interface Sci 1985;108:215–25.
[279] Imae T, Ikeda S. Sphere-rod transition of micelles of tetradecyltrimethylammonium halides in aqueous sodium halide solutions and flexibility and
entanglement of long rodlike micelles. J Phys Chem 1986;90:5216–23.
[280] Nomura H, Koda S, Matsuoka T, Hiyama T, Shibata R, Kato S. Study of salt effects on the micelle-monomer exchange process of octyl-, decyl-, and
dodecyltrimethylammonium bromide in aqueous solutions by means of ultrasonic relaxation spectroscopy. J Colloid Interface Sci 2000;230:22–8.
[281] Weican Z, Ganzuo L, Jianhai M, Qiang S, Liqiang Z, Haojun L, et al. Chin Sci Bull 2000;45:1854.
[282] Khatory A, Lequeux F, Kern F, Candau SJ. Linear and nonlinear viscoelasticity of semidilute solutions of wormlike micelles at high salt content.
Langmuir 1993;9:1456–64.
[283] Magid L, Han Z, Li Z, Butler P. Tuning microstructure of cationic micelles on multiple length scales: the role of electrostatics and specific ion binding.
Langmuir 2000;16:149–56.
[284] Clint JH. Micellization of mixed nonionic surface active agents. J Chem Soc, Faraday Trans 1975;1(71):1327–34.
[285] Holland PM, Rubingh DN. Nonideal multicomponent mixed micelle model. J Phys Chem 1983;87:1984–90.
[286] Burman AD, Dey T, Mukherjee B, Das AR. Solution properties of the binary and ternary combination of sodium dodecyl benzene sulfonate,
polyoxyethylene sorbitan monlaurate, and polyoxyethylene lauryl ether. Langmuir 2000;16:10020–7.
[287] Dar AA, Rather GM, Ghosh S, Das AR. Micellization and interfacial behavior of binary and ternary mixtures of model cationic and nonionic surfactants
in aqueous NaCl medium. J Colloid Interface Sci 2008;322:572–81.
[288] Preiss UP, Eiden P, Luczak J, Jungnickel C. Modeling the influence of salts on the critical micelle concentration of ionic surfactants. J Colloid Interface
Sci 2013;412:13–6.
[289] Shinoda K. The critical micelle concentration of soap mixtures (two component mixture). J Phys Chem 1954;58:541–4.
[290] Shinoda K, Nakagawa T. Colloidal surfactants. New York: Academic Press; 1963.
[291] Kameyama K, Muroya A, Takagi T. Properties of a mixed micellar system of sodium dodecyl sulfate and octylglucoside. J Colloid Interface Sci
1997;196:48–52.
[292] Growcock FB, Frenier WW. Kinetics of steel corrosion in hydrochloric acid inhibited with trans-cinnamaldehyde. J Electrochem Soc 1988;135:817–23.
[293] Endo S, Goss K. Predicting partition coefficients of polyfluorinated and organosilicon compounds using polyparameter linear free energy
relationships. Environ Sci Technol 2014;48:2776–84.
[294] Ghoulam MB, Moatadid N, Graciaa A, Lachaise J. Quantitative effect of nonionic surfactant partitioning on the hydrophile-lipophile balance
temperature. Langmuir 2004;20:2584–9.
[295] Graciaa A, Andérez J, Bracho C, Lachaise J, Salager J, Tolosa L, et al. The selective partitioning of the oligomers of polyethoxylated surfactant mixtures
between interface and oil and water bulk phases. Adv Coll Interface Sci 2006;123–126:63–73.
[296] Salager J, Marquez N, Graciaa A, Lachaise J. Partitioning of ethoxylated octylphenols surfactants in microemulsion-oil-water systems: influence of
temperature and relation between partitioning coefficient and physicochemical formulation. Langmuir 2000;16:5534–9.
[297] Gomez del Rio J, Hayes D, Urban VS. Partitioning behavior of an acid-cleavable, 1, 3-dioxolane alkyl ethoxylate, surfactant in single and binary
surfactant mixtures for 2- and 3-phase microemulsion systems according to ethoxylate and head group size. J Colloid Interface Sci 2010;352:424–35.
[298] Crook EH, Fordyce DB, Trebbi GF. Partition coefficients of normal distribution and homogeneous p,t-octylphenoxyethoxyeth. J Colloid Interface Sci
1965;20:191–204.
[299] Brooks BW, Richmond HN. The application of a mixed nonionic surfactant theory to transitional emulsion phase inversion: 1. derivation of a mixed
surfactant partitioning model. J Colloid Interface Sci 1994;162:59–66.
[300] Brooks BW, Richmond HN. The application of a mixed nonionic surfactant theory to transitional emulsion phase inversion: 2. the relationship
between surfactant partitioning and transitional inversion-a thermodynamic treatment. J Colloid Interface Sci 1994;162:67–74.
[301] Salager JL, Marquez N, Anton RE, Graciaa A, Lachaise J. Retrograde transition in the phase behavior of surfactant-oil-water systems produced by an
alcohol scan. Langmuir 1995;11:37–41.
[302] Ravera F, Ferrari M, Liggieri L, Miller R, Passerone A. Measurement of the partition coefficient of surfactants in water/oil systems. Langmuir
1997;13:4817–20.
[303] Ghoulam B, Moatadid N, Graciaa A, Lachaise J. Effects of oxyethylene chain length and temperature on partitioning of homologous polyoxyethylene
nonionic surfactants between water and isooctane. Langmuir 2002;18:4367–71.
[304] Zhu Y, Molinier V, Durand M, Lavergne A, Aubry J. Amphiphilic properties of hydrotropes derived from isosorbide: endo/exo isomeric and
temperature dependence. Langmuir 2009;25:13419–25.
[305] Van de Voorde A, Lorgeous C, Gromaire M, Chebbo G. Analysis of quaternary ammonium compounds in urban storm water samples. Environ Pollut
2012;164:150–7.
[306] Warr GG, Grieser F, Healy TW. Distribution of polydisperse nonionic surfactants between oil and water. J Phys Chem 1983;87:4520–4.
[307] Cowell MA, Kibbey TCG, Zimmerman JB, Hayes KF. Partitioning of ethoxylated nonionic surfactants in water/NAPL systems: effects of surfactants and
NAPL properties. Environ Sci Technol 2000;34:1583–8.
[308] Graciaa A, Lachaise J, Sayous JG, Grenier P, Yiv S, Schechter RS, et al. The partitioning of complex surfactant mixtures between oil/water/
microemulsion phases at high surfactant concentrations. J Colloid Interface Sci 1983;93:474–86.
[309] Balcan M, Anghel D. The partition of ethoxylated non-ionic surfactants between two non-miscible phases. Colloid Polym Sci 2005;283:982–6.
[310] Pradines V, Despous S, Claparols C, Martins N, Micheau J, Lavabre D, et al. Partition of dissociable compounds in two-phase liquid systems: a
theoretical and experimental study. J Phys Org Chem 2006;19:350–8.
[311] Harusawa F, Saito T, Nakajima H, Fukushima S. Partition isotherms of nonionic surfactants in the water-cyclohexane system and the type of emulsion
produced. J Colloid Interface Sci 1980;74:435–40.
[312] Alaei P, Binks BP, Fletcher PDI. Surfactant properties of alkyl benzyl dimethyl ammonium chloride oilfield corrosion inhibitors. In: NACE corrosion
conference; 2013. Paper no. 2158.
[313] Aveyard R, Binks BP, Clark S, Fletcher PDI. Effects of temperature on the partitioning and adsorption of C12E5 in heptane–water mixtures. J Chem Soc,
Faraday Trans 1990;1(86):3111–5.
[314] Aveyard R, Binks BP, Clark S, Mead J. Interfacial tension minima in oil–water–surfactant systems. Behaviour of alkane–aqueous NaCl systems
containing aerosol OT. J Chem Soc, Faraday Trans 1986;1(82):125–42.
[315] Marenich AV, Cramer CJ, Truhlar DG. Universal solvation model based on solute electron density and on a continuum model of the solvent defined by
the bulk dielectric constant and atomic surface tensions. J Phys Chem B 2009;113:6378–96.
[316] Scalmani G, Frisch MJ. Continuous surface charge polarizable continuum models of solvation. I. General formalism. J Chem Phys 2010;132:114110–5.
[317] Taylor CD. Design and prediction of corrosion inhibitors from quantum chemistry. I. Can effective partition coefficients for organic corrosion
inhibitors be determined from first-principles? J Electrochem Soc 2015;162:C340–6.
[318] Taylor CD, Chandra A, Vera J, Sridhara N. Design and prediction of corrosion inhibitors from quantum chemistry. II. A general framework for
prediction of effective oil/water partition coefficients and speciation from quantum chemistry. J Electrochem Soc 2015;162:C347–53.
[319] Ellegaard MD, Abildskov J, O’Connell JP. Method for predicting solubilities of solids in mixed solvents. AIChE J 2009;55:1256–64.
[320] Smith PE, Mazo RM. On the theory of solute solubility in mixed solvents. J Phys Chem B 2008;112:7875–84.
[321] Van der Vegt NFA, van Gunsteren WF. Entropic contributions in cosolvent binding to hydrophobic solutes in water. J Phys Chem B 2004;108:1056–64.
[322] Marcus Y. Solubility and solvation in mixed solvent systems. Pure Appl Chem 2009;62:2069–76.
[323] Williams NA, Amidon GL. The estimation of solubility in binary solvents: application of the reduced 3-suffix solubility equation to ethanol-water
mixtures. Pharm Res 1988;5:193–5.
[324] Chamberlin AC, Levitt DG, Cramer CJ, Truhlar DG. Modeling free energies of solvation in olive oil. Mol Pharm 2008;5:1064–79.
[325] James AD, Wates JM, Jones EW. Determination of the hydrophilicities of nitrogen-based surfactants by measurement of partition coefficients between
heptane and water. J Colloid Interface Sci 1993;160:158–65.
[326] Klopman G, Zhu H. Estimation of the aqueous solubility of organic molecules by the group contribution approach. J Chem Inf Comput Sci
2001;41:439–45.
[327] Harusawa F, Tanaka M. Mixed micelle formation in two-phase systems. J Phys Chem 1981;85:882–5.
[328] Free ML, Miller JD. The significance of collector colloid adsorption phenomena in the fluorite/oleate system. Int J Min Proc 1996;48:197–216.
[329] van Duin ACT, Dasgupta S, Lorant F, Goddard III WA. ReaxFF: a reactive force field for hydrocarbons. J Phys Chem A 2001;105:9396–409.
[330] Raju M, Kim SY, van Duin ACT, Fichthorn KA. ReaxFF reactive force field study of the dissociation of water on titania surfaces. J Phys Chem C
2013;117:10558–72.
[331] Selvam P, Tsuboi H, Koyama M, Kubo M, Miyamoto A. Tight-binding quantum chemical molecular dynamics method: a novel approach to the
understanding and design of new materials and catalysts. Catal Today 2005;100:11–25.
[332] Valentini P, Schwartzentruber TE, Cozmuta I. ReaxFF grand canonical Monte Carlo simulation of adsorption and dissociation of oxygen on platinum
(1 1 1). Surf Sci 2011;605:1941–50.
[333] Hong T, Jepson WP. Corrosion inhibitor studies in large flow loop at high temperature and high pressure. Corros Sci 2001;43:1839–49.
[334] Barmatov E, Hughes T, Nagl M. Efficiency of film-forming corrosion inhibitors in strong hydrochloric acid under laminar and turbulent flow
conditions. Corros Sci 2015;92:85–94.
[335] Xu LY, Cheng YF. Development of a finite element model for simulation and prediction of mechanoelectrochemical effect of pipeline corrosion. Corros
Sci 2013;73:150–60.
[336] Macdonald DD. On the existence of our metals-based civilization I. phase-space analysis. J Electrochem Soc 2006;153:B213–24.
[337] Chao CY, Lin LF, Macdonald DD. A point defect model for anodic passive films I. film growth kinetics. J Electrochem Soc 1981;128:1187–94.
[338] Lin LF, Chao CY, Macdonald DD. Chemical breakdown and pit initiation. J Electrochem Soc 1981;128:1194–8.
[339] Frenier WW. US patent: 5,096,618. Dowell Schlumberger Incorporated, Tulsa, Okla; 1992.
[340] Papavinasam S, Revie RW, Attard M, Demoz A, Michaelian K. Comparison of techniques for monitoring corrosion inhibitors in oil and gas pipelines.
Corrosion 2003;59:1096–111.
[341] Paty BB, Singh DDN. Solvents’ role on HCl-induced corrosion of mild steel: its control by propargyl alcohol and metal cations. Corrosion
1992;48:442–6.
[342] G1-03 standard practice for preparing, cleaning, and evaluating corrosion test specimens. ASTM International; 2003.
[343] Finšgar M. 2-Mercaptobenzimidazole as a copper corrosion inhibitor: part I. Long-term immersion, 3D-profilometry, and electrochemistry. Corros Sci
2013;72:82–9.
[344] Finšgar M. 2-mercaptobenzimidazole as a copper corrosion inhibitor: part II. Surface analysis using X-ray photoelectron spectroscopy. Corros Sci
2013;72:90–8.
[345] Finšgar M. Galvanic series of different stainless steels and copper- and aluminium-based materials in acid solutions. Corros Sci 2013;68:51–6.
[346] Saassouh B, Lounis Z. Probabilistic modeling of chloride-induced corrosion in concrete structures using first- and second-order reliability methods.
Cement Concr Compos 2012;34:1082–93.
[347] Vasyliev GS. The influence of flow rate on corrosion of mild steel in hot tap water. Corros Sci 2015;98:33–9.
View publication stats

A Review of Surfactant As Corrosion Inhibitors

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A Review of Surfactant As Corrosion Inhibitors

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

A Review of Surfactants as Corrosion Inhibitors and Associated Modeling

Article in Progress in Materials Science · July 2017

Yakun Zhu Michael L. Free

SEE PROFILE SEE PROFILE

A novel Ti metal production approach View project

Lithium isotope separation View project

The user has requested enhancement of the downloaded file.

Contents lists available at ScienceDirect

Progress in Materials Science

A review of surfactants as corrosion inhibitors and associated

5.2. Diffusion for slow moving or steady-state systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

AFM atomic force microscopy

MQSAR modified quantitative structure activity relation

Ssn synergistic parameter

2. Corrosion background information

2.1. Corrosion electrochemistry

Net reaction: Fe + 2H+ = Fe2+ + H2

Iron oxide barrier layer

2.2. CO2 corrosion

Fe $ Fe2þ þ 2e ð3Þ

Fe2þ þ 2HCO3 $ FeðHCO3 Þ2 ð5Þ

FeðHCO3 Þ2 $ FeCO3 þ CO2 þ H2 O ð6Þ

2HCO3 þ 2e $ H2 þ 2CO2

½FeðOHÞ2 ads þ ½H2 CO3 ads $ FeCO3 þ 2H2 O ð11Þ

½FeClðOHÞ ads $ FeClOH þ e ð14Þ

2.3. H2S corrosion

FeHS $ FeHS þ e ð17Þ

FeHS $ FeHSþ þ e ð18Þ

FeHSþ $ FeS þ Hþ ð19Þ

2.4. Corrosion reaction rate theory

Corrosion is driven by electrochemical reactions. For an electrochemical reaction given as:

The analogous expressions for the cathodic reaction are:

iocorr (low flow)

iocorr (high flow)

5.1. Diffusion for short-time start up intervals

5.2. Diffusion for slow moving or steady-state systems

U 1=2 mk1=6 D1=3

U 9=10 mk17=30 D1=3

flr ¼ 1:6x1=2 m1=6

where x is the angular velocity.

where dp is the diameter of the particle.

5.3. Identifying mass transport control

5.4. Combination of diffusion and Butler-Volmer kinetics

6.1. Introduction to surfactant

6.2. Hydrophilicity and hydrophobicity of surfactant

Surfactant name Structure Symbol n

n-Benzalkonium chloride (BAC) Cn, CnCl, or 12, 14,

Polyoxythylene alkyl ether CnEn0 12, 16

n-Alkyltrimethyl ammonium surfactant CnTAX 10, 12,

Potassium alkanoate CnH(2n+1)KO2 9, 11

Dodecylpyridinium bromide DDPB 12

Sodium dodecyl sulfate SDS 12

Sodium bis-(2-ethylhexyl)- AOT –

Cetylpyridinium chloride CPC 14

Surfactant name Structure Symbol n

Polyethylene glycol esters – –

N-based alkyl amines and derivatives CnNH2 6

air air air air

solution solution solution solution

substrate substrate substrate substrate

specific adsorption/ isolated aggregate widespread aggregates on solution aggregation/

6.3. Adsorption mechanism