Professional Documents
Culture Documents
Raoul Zana
Institut C.Sadron (CNRS-ULP)
Strasbourg, France
Jiding Xia
Wuxi University of Light Industry
Wuxi, People’s Republic of China
ADVISORY BOARD
DANIEL BLANKSCHTEIN
Department of Chemical Engineering
Massachusetts Institute of Technology
Cambridge, Massachusetts
S.KARABORNI
Shell International Petroleum
Company Limited
London, England
LISA B.QUENCER
The Dow Chemical Company
Midland, Michigan
JOHN F.SCAMEHORN
Institute for Applied Surfactant
Research
University of Oklahoma
Norman, Oklahoma
P.SOMASUNDARAN
Henry Krumb School of Mines
Columbia University
New York, New York
ERIC W.KALER
Department of Chemical Engineering
University of Delaware
Newark, Delaware
CLARENCE MILLER
Department of Chemical Engineering
Rice University
Houston, Texas
DON RUBINGH
The Procter & Gamble Company
Cincinnati, Ohio
BEREND SMIT
Shell International Oil Products B.V.
Amsterdam, The Netherlands
JOHN TEXTER
Strider Research Corporation
Rochester, New York
1. Nonionic Surfactants, edited by Martin J.Schick (see also Volumes 19, 23, and 60)
2. Solvent Properties of Surfactant Solutions, edited by Kozo Shinoda (see Volume 55)
4. Cationic Surfactants, edited by Eric Jungermann (see also Volumes 34, 37, and 53)
5. Detergency: Theory and Test Methods (in three parts), edited by W.G. Cutler and
R.C.Davis (see also Volume 20)
6. Emulsions and Emulsion Technology (in three parts), edited by Kenneth J. Lissant
7. Anionic Surfactants (in two parts), edited by Warner M.Linfield (see Volume 56)
12. Amphoteric Surfactants, edited by B.R.Bluestein and Clifford L Hilton (see Volume
59)
13. Demulsification: Industrial Applications, Kenneth J.Lissant
17. Interfacial Phenomena: Equilibrium and Dynamic Effects, Clarence A.Miller and
P.Neogi
18. Surfactant Biodegradation: Second Edition, Revised and Expanded, R.D. Swisher
20. Detergency: Theory and Technology, edited by W.Gale Cutler and Erik Kissa
25. Biosurfactants and Biotechnology, edited by Naim Kosaric, W.L.Cairns, and Neil
C.C.Gray
31. Crystallization and Polymorphism of Fats and Fatty Acids, edited by Nissim Garti
and Kiyotaka Sato
32. Interfacial Phenomena in Coal Technology, edited by Gregory D.Botsaris and Yuli
M.Glazman
35. Alkylene Oxides and Their Polymers, F.E.Bailey, Jr., and Joseph V. Koleske
37. Cationic Surfactants: Physical Chemistry, edited by Donn N, Rubingh and Paul
M.Holland
44. Organized Solutions: Surfactants in Science and Technology, edited by Stig E.Friberg
and Björn Lindman
48. Mixed Surfactant Systems, edited by Keizo Ogino and Masahiko Abe
47. Coagulation and Flocculation: Theory and Applications, edited by Bohuslav Dobiáš
53. Cationic Surfactants: Analytical and Bioiogical Evaluation, edited by John Cross and
Edward J.Singer
68. Surfactants in Cosmetics: Second Edition, Revised and Expanded, edited by Martin
M.Rieger and Linda D.Rhein
69. Enzymes in Detergency, edited by Jan H.van Ee, Onno Misset, and Erik J. Baas
73. Anionic Surfactants: Analytical Chemistry, Second Edition, Revised and Expanded,
edited by John Cross
78. Surfaces of Nanoparticles and Porous Materials, edited by James A. Schwarz and
Cristian I.Contescu
81. Solid-Liquid Dispersions, Bohuslav Dobiáš, Xueping Qiu, and Wolfgang von
Rybinski
82. Handbook of Detergents, editor in chief: Uri Zoller Part A: Properties, edited by Guy
Broze
85. Interfacial Forces and Fields: Theory and Applications, edited by Jyh-Ping Hsu
91. Nonionic Surfactants: Alkyl Polyglucosides, edited by Dieter Balzer and Harald
Lüders
94. Surface Characteristics of Fibers and Textiles, edited by Christopher M. Pastore and
Paul Kiekens
96. Analysis of Surfactants: Second Edition, Revised and Expanded, Thomas M. Schmitt
97. Fluorinated Surfactants and Repellents: Second Edition, Revised and Expanded, Erik
Kissa
105. Colloid and Surface Properties of Clays and Related Minerals, Rossman F. Giese
and Carel J.van Oss
110. Biopolymers at Interfaces: Second Edition, Revised and Expanded, edited by Martin
Malmsten
111. Biomolecular Films: Design, Function, and Applications, edited by James F. Rusiing
113. Liquid Interfacial Systems: Oscillations and instability, Rudolph V.Birikh, Vladimir
A.Briskman, Manuel G.Velarde, and Jean-Claude Legros
Surfactants are present in most aspects of our daily life: in the soap or gel we use when
showering in the morning, in the creams and cosmetics we spread on our skin, in many of
the foods we have from breakfast to night, in the detergents we use in our washing
machine and dishwasher, and in some of the medications we may be prescribed. They are
also present in the processes used to prepare the clothes we wear, and the metals and
plastics in most of the tools, appliances, cars, and so forth, we use. It should therefore not
come as a surprise that research on, and developments of, new surfactants has always
been very active in spite of the fact that surfactant catalogs list hundreds if not thousands
of surfactants of all types—anionic, cationic, zwitterionic, and nonionic. The world
production of surfactants is probably close to 10 million tons per year, worth several
billions of dollars. Any new surfactant with novel properties or improved performances
or capable of improving the economics of a given process would translate into savings of
millions of dollars. In addition, new surfactants with lower toxicity or environment-
friendly characteristics have a promising future, as new regulations that will most likely
be enacted in the coming years will require the surfactants used in formulations to have a
lower level of toxicity and less impact on the environment and to be more easily
degraded by natural biological processes.
A new type of surfactant, the so-called dimeric or gemini surfactants, has recently
generated much interest in academic circles and among scientists at surfactant-producing
companies. These surfactants are made up of two amphiphilic surfactant-like moieties
connected at the level of the head groups or very close to the head groups by a spacer
group of varied nature. Several reasons can be given for the present interest in gemini
surfactants. First their critical micellization concentration (cmc) is generally at least one
order of magnitude lower than that of the corresponding conventional (monomeric)
surfactants. Second, they are 10–100 times more efficient at reducing the surface tension
of water and the interfacial tension of the oil-water interface than conventional
surfactants. Third, the aqueous solutions of some dimeric surfactants with a short spacer
can have extremely interesting rheological properties (viscoelasticity, shear thickening).
Finally, the microstructure of some gemini surfactant solutions shows unexpected but
remarkable micellar shapes. In fact, the properties of gemini surfactants are in some
respects so unexpected and superior to those of comparable conventional surfactants that
Rosen, in a paper published in Chemtech in March 1993, referred to gemini surfactants as
“surfactants for the nineties.” This statement was premature, but certainly gemini
surfactants are the surfactants for the future.
Gemini surfactants can be generated with an enormous variety of structures because it
is in principle possible to take any two identical or different surfactants among the
available ones and connect them by a spacer group that can be hydrophilic or
hydrophobic, flexible or rigid, heteroatomic, aromatic, and so on. New properties will
probably be discovered for the gemini surfactants that can be synthesized. In addition, the
concept of gemini (dimeric) surfactants has been extended to longer homologs, trimeric
surfactants made up of three surfactant-like moieties connected by two spacer groups,
tetrameric surfactants, etc. The number of possible structures is truly mind-boggling.
Research on gemini surfactants is expanding, both in universities and in chemical
companies. Formulations based on anionic dimeric surfactants made by Condea (now
Sasol GmbH, Marl, Germany) have reached the market. Because of their superior
properties and special performances, gemini surfactants have been suggested in
formulating products for more effective wetting, solubilizing, dispersing, thickening,
making of microemulsion, antimicrobial activity, and other industrial uses. It thus
appeared timely to prepare a volume that would present the state of the art of gemini
surfactants for those already doing research work on these surfactants or planning to do
so, in both the academic and industrial worlds.
This volume comprises 13 chapters that provide as complete a view of gemini
surfactants as possible at the time of the writing of the book. It is hoped that the
interesting properties that these surfactants display will draw more people in the field.
The help of synthetic organic chemists is much needed for the complicated and
challenging synthesis involved in generating gemini surfactants with complex structure.
Physicochemists and physicists will follow to study the properties of these new
compounds. Industrial surfactant scientists will accompany these developments at all
stages.
Raoul Zana
Jiding Xia
Contents
Preface xii
Contributors xvi
1. Introduction 1
Raoul Zana and Jiding Xia
2. Synthesis of Gemini (Dimeric) and Related Surfactants 9
Isao Ikeda
3. Models of Gemini Surfactants 37
Haim Diamant and David Andelman
4. Adsorption and Surface Tension Behavior of Gemini Surfactants at Air- 64
Water, Oil-Water, and Solid-Water Interfaces
Elias I.Franses, Maria Rosa Infante, Lourdes Pérez, Aurora Pinazo, and
Alissa J.Prosser
5. State of Gemini Surfactants in Solution at Concentrations Below the cmc 92
Raoul Zana
6. Gemini (Dimeric) Surfactants in Water: Solubility, cmc, Thermodynamics 108
of Micellization, and Interaction with Water-Soluble Polymers
Raoul Zana
7. Properties of Micelles and of Micellar Solutions of Gemini (Dimeric) 139
Surfactants
Raoul Zana
8. Rheology of Solutions of Gemini Surfactants 183
Martin In
9. Phase Behavior of Gemini Surfactants 210
Raoul Zana and Martin In
10. Mixed Micellization Between Dimeric (Gemini) Surfactants and 231
Conventional Surfactants
Raoul Zana and Jiding Xia
11. Special Gemini Surfactants: Nonionic, Zwitterionic, Fluorinated, and Amino 251
Acid Based
Tim W.Davey
12. Structure-Performance Relationships in Gemini Surfactants 277
Yun-Peng Zhu
13. Applications of Gemini Surfactants 296
Jiding Xia and Raoul Zana
Index 316
Contributors
David Andelman Beverly and Raymond Sackler Faculty of Exact Sciences, School of
Physics and Astronomy, Tel Aviv University, Tel Aviv, Israel
Tim W.Davey Polymer Technology, Dulux, Melbourne, Victoria, Australia
Haim Diamant Beverly and Raymond Sackler Faculty of Exact Sciences, School of
Chemistry, Tel Aviv University, Tel Aviv, Israel
Elias I.Franses School of Chemical Engineering, Purdue University, West Lafayette,
Indiana, U.S.A.
Isao Ikeda Professor Emeritus, Osaka University, Osaka, Japan
Martin In Groupe de Dynamique des Phases Condensées, CNRS-Université Montpellier
2, Montpellier, France
Maria Rosa Infante Instituto de Investigaciones Químicas y Ambientals de Barcelona,
Consejo Superior de Investigaciones Cientificas, Barcelona, Spain
Lourdes Pérez Instituto de Investigaciones Químicas y Ambientals de Barcelona,
Consejo Superior de Investigaciones Cientificas, Barcelona, Spain
Aurora Pinazo Instituto de Investigaciones Químicas y Ambientals de Barcelona,
Consejo Superior de Investigaciones Cientificas, Barcelona, Spain
Alissa J.Prosser School of Chemical Engineering, Purdue University, West Lafayette,
Indiana, U.S.A.
Jiding Xia Department of Chemical Engineering, Wuxi University of Light Industry,
Wuxi, People’s Republic of China
Raoul Zana Institut C.Sadron (CNRS-ULP), Strasbourg, France
Yun-Peng Zhu Edgewater Laboratory, Unilever HPC, Edgewater, New
1
Introduction
RAOUL ZANA Institut C.Sadron (CNRS-ULP), Strasbourg, France
JIDING XIA Wuxi University of Light Industry, Wuxi, People’s
Republic of China
I.
DEFINITION OF GEMINI (DIMERIC) SURFACTANTS AND
HISTORIC ASPECTS
Gemini (also called dimeric) surfactants represent a new class of surfactants made up of
two identical or different amphiphilic moieties having the structure of conventional
(monomeric) surfactants connected by a spacer group [1]. The spacer may be
hydrophobic (aliphatic or aromatic) or hydrophilic (polyether), short (two methylene
groups) or long (up to 20 and more methylene groups), rigid (stilbene) or flexible
(polymethylene chain). Figure la gives a schematic representation of a gemini surfactant.
At the outset, it must be emphasized that the spacer group must connect the two
amphiphilic moieties at the level of, or in close vicinity to, the head groups. If the
connection takes place in the middle or toward the end of the alkyl chains of the
amphiphilic moieties as in Fig. 1b, the surfactant is then a bolaform surfactant endowed
with properties that are inferior to those of conventional surfactants.
Gemini surfactants have been known in the patent literature since 1935. This literature
has been recently reviewed [2,3] and is not dealt with here. To the best of our knowledge
the first report on gemini surfactants in the scientific literature is that of Bunton et al. in
1971 [4]. These authors synthesized bisquaternary ammonium bromide gemini
surfactants and studied how the micelles of these surfactants affect the rate of chemical
reactions. This work was followed by that of Devinsky et al. [5] who synthesized
bisquaternary ammonium gemini surfactants with a great variety of structures, and that of
Okahara et al. [6], who synthesized a large number of anionic gemini surfactants. The
variety of gemini surfactants that have been synthesized to this day is already enormous.
It includes anionic, cationic, zwitterionic, and non-ionic surfactants with all kinds of
spacer group. Gemini surfactants with al-
Gemini surfactants 2
homologs that are referred to in this volume as oligomeric surfactants [1]. Figure 1c
shows a trimeric surfactant. A limited number of trimeric and tetrameric surfactants have
been synthesized [12,13]. Their properties were found to be superior to those of the
corresponding dimeric surfactants. In addition, these surfactants are intermediate between
conventional surfactants and polymeric surfactants (i.e., polymers with a repeat unit that
is amphiphilic). The systematic study of oligomeric surfactants may help in
understanding some properties of polymeric surfactants.
II.
INTEREST OF GEMINI SURFACTANTS
There are several reasons for the current interest in gemini surfactants in both the
academic and the industrial circles working on surfactants.
1. Gemini surfactants are characterized by cmc’s that are one to two orders of magnitude
lower than for the corresponding conventional (monomeric) surfactants. For instance,
Fig. 2 shows that the cmc of the dimeric surfactant 12–2–12 [dimethylene-1,2-
bis(dodecyldimethylammonium
III.
CHAPTER OVERVIEW
REFERENCES
18. Molinaro, A.; Suilipo, A.; Lanzatta, R.; Parrilli, M.; Malvagna, P.; Evidente, A.; Surico, G. Eur.
J. Org. Chem. 2002, 3119.
2
Synthesis of Gemini (Dimeric) and Related
Surfactants
ISAO IKEDA Osaka University, Osaka, Japan
I.
INTRODUCTION
Gemini (dimeric) surfactants are defined as surfactants that are made up of two
amphiphilic moieties connected at the level of, or close to, the head groups by a spacer
group. In the past 15 years, many different types of gemini surfactant have been
synthesized due to their unique properties in aqueous solution. At first, however, studies
on the catalysis of chemical reactions by micelles of cationic dimeric surfactants and on
the use of these surfactants as bactericidal and fungicidal agents were reported some 30
years ago. It might have been recognized at that time that cationic surfactants showed
unique micelle-forming and surface-adsorbing properties when connecting them two by
two to generate dimeric surfactants. Thus, in 1971, Bunton et al. [1] reported the
synthesis of dimeric cationic surfactants with an alkylene or 2-butynylene spacer group
and their use in micelle catalysis. In 1985–1986, Devinsky et al. [2,3] reported on the
relationship between the structure and surface activity of four series of cationic dimeric
surfactants and on their high antimicrobial activity. Later, unique properties concerning
not only surface activity but also molecular aggregation were evidenced. Currently, the
relationship between structure and properties of dimeric surfactants has been fully
evidenced for anionic as well as cationic gemini surfactants. The activity in the synthesis
of these surfactants has shifted from simple gemini surfactants to multi-armed gemini,
polyionic gemini, trimeric, tetrameric, and oligomeric surfactants, as well as in
developing new kinds of hydrophilic group, hydrophobic chain, spacer, and counterion.
Recently, gemini surfactants have attracted much attention as potential agents in gene
therapy, and their properties as vehicle for gene delivery into cells (transfection) were
reported.
Reviews concerning the synthesis and structures of gemini surfactants have been
published [4–8].
Gemini surfactants 10
II.
CATIONIC GEMINI SURFACTANTS
A.
General Synthesis
In general, cationic gemini surfactants can be synthesized by heating under reflux
alcoholic (ethanol) solutions of alkyldimethylamine and α,ω-alkylene-dihalide or of
N,N,N′,N′-tetramethyl-α,ω-alkylenediamine and alkyl halide in order to achieve the
quaternization of the amine (Scheme 1, reactions 1 and 2, respectively) [1–3,9–23]. These
surfactants can also be obtained by amida-
tion of dibasic acid dichloride by alkylamine into diamide, followed by reduction with
LiAlH4 and quaternization with a methylation reagent such as methyl halide and
dimethylsulfate (Scheme 2, reaction 3).
When using reactions 1 and 2 with an aliphatic halide or dihalide as the starting
material [1–3,9–14,20–23], the quaternization reaction may require a few days to reach
completion and to yield the end product with high purity, even if an excess of the
monofunctional starting compound is used. This is due to the low reactivity of the
aliphatic halide. Especially with short alkylene tetramethylethylenediamine, it has been
reported that the reaction mixture must be heated for up to 12 days for the quaternization
reaction of the two dimethylamino groups to be completed [1]. The reaction time can be
shortened by replacing ethanol [boiling point (bp): 78°C] by 2-propanol (bp: 82°C) [24]
or 1-propanol (bp: 97°C) [25] as a solvent. Also, sodiumiodide can be added as a catalyst.
However, this results in the contamination of the final product by the iodide ion.
As shown in Scheme 1, 4-alkylpyridine [22] and 4,4’-connected bipyridine [23] can be
used as a starting material for the synthesis of gemini surfactants with pyridinium halide
head groups.
Because the tendency of aliphatic cationic gemini surfactants to crystallize is
sometimes not high enough, purification of the end product by recrystallization in order
to eliminate a monocationic intermediate may not be effective. This is demonstrated by
the few satisfactory elemental analysis data reported when using this procedure. One
study that reported a successful
time, increasing the reaction temperature, and using a larger excess amount of
monofunctional alkyldimethylamine or alkyl halide in reactions 1 and 2 (Scheme 1),
respectively. The procedure can be improved by following the course of the reaction by
determining the tertiary amine used in excess. This allows one to know when the reaction
has reached completion within the accuracy of the analysis method adopted. Gemini
surfactants with higher purity can be expected [26].
Despite the problems discussed earlier, reactions 1 and 2 in Scheme 1 have been much
used for the preparation of cationic gemini surfactants due to the simple synthetic
procedures involved.
Cationic gemini surfactants composed of perdeuterododecyldimethylammonio groups
and a perdeuteroalkylene (propylene, butylene, hexylene, and dodecylene) or
perhydroxylylene spacer were synthesized by this method for neutron reflectivity study
[27].
B.
Synthesis from More Reactive Halide or Dihalide Raw Materials
In the case of more reactive halides such as benzylic [15–17,21], allylic [18], and
propargylic [1,18,19] halides as the starting material, the problem concerning the reaction
time and, thus, purification becomes less acute. Many cationic gemini surfactants with
various types of spacer as well as typical alkylene spacers have been reported, as shown
in Scheme 1.
C.
Alternative Stepwise Synthesis to Purer Products
When a more purified product is required, synthesis via reaction 3 (Scheme 2) is
recommended on the basis of synthetic organic chemistry. Indeed, the intermediate
obtained in each reaction step that proceeds quickly can be purified thoroughly by
column chromatography. Also, the final quaternization process with an efficient
methylation reagent such as methyl halide or dimethyl sulfate proceeds usually much
faster and affords a purer product.
D.
Synthesis of Gemini Surfactants Using Epichlorohydrin
If the 2-hydroxy-1,3-propylene moiety is desired as the spacer, epichlorohydrin is a more
preferable material than 1,3-dichloro-2-propanol to obtain the same gemini surfactant
(see Scheme 3). Indeed, the reaction time is much reduced when using epichlorohydrin
[10]. In this case, an amount of hydrogen chloride equivalent to half the amount of amine
should be added to the reaction solution before heating [10,28].
Diglycidyl compounds derived from epichlorohydrin are potential starting materials
for the synthesis of cationic gemini surfactants, as shown in Scheme 4 [29].
E.
Synthesis of Dissymmetric Cationic Gemini Surfactants
A series of dissymmetric gemini surfactants was synthesized by utilizing the difference
between reaction rates of two steps of quaternization reactions in reaction 2 of Scheme 1.
Thus, a less reactive starting compound, such as aliphatic bromide and chloride, in a
smaller quantity than the stoichiometry yields the monoammonium halide under mild
conditions, avoiding the formation of diammonium dihalide. This monoammonium salt
can be purified with ease by extracting the unreacted starting materials. Any kind of
halide can be used in the second quaternization step that quantitatively yields
dissymmetric gemini surfactants by the use of required conditions to complete the
reaction as shown in Scheme 5 [30,31].
SCHEME 5 Synthesis of
dissymmetric gemini surfactants.
(From Refs. 30 and 31.)
F.
Synthesis of Chiral Cationic Gemini Surfactants
A chiral gemini surfactant with two methoxy substituents, which might show a specific
molecular aggregation behavior due to its symmetric structure, was synthesized as shown
in Scheme 6 [32]. It forms very large aggregates in aqueous solution, in which the
kinetics of a micelle-catalyzed intramolecular SN2 reaction was studied. Along with the
kinetics, the order of substrate-surfactant aggregation was discussed and the results for
the 2,3-dimethoxy-bisammoniobutane and the bisammoniobutane surfactants were
compared [32].
G.
Synthesis from Amino Acids and Spermine
Gemini surfactants derived from amino acids are chiral. Amino acids are potential
starting compounds as a pivot-forming compound connecting a lipophilic chain, a
hydrophilic group, and a spacer, as well as a component of hydrophilic moiety, as shown
in Scheme 7 [33]. Lysine and 2,4-diaminobutyric acid form polycationic gemini
surfactants, as shown in Scheme 8, by relatively simple synthesis using standard peptide
Gemini surfactants 16
H.
Synthesis of Cationic Gemini Surfactants with Sugar Moiety
A disaccharide can form chiral gemini surfactants where the spacer bears many hydroxyl
groups, as shown in Scheme 9 [38]. Sugars can also constitute the hydrophilic head
groups in cationic and nonionic gemini surfactants, as will be shown in Scheme 27 [39].
I.
Cationic Gemini Surfactants with Chiral Counterion
Chiral cationic gemini surfactants where the chiral center is located only in the
counterion can be prepared by ion exchange (Scheme 10) from gemini surfactants
obtained using reaction 2 [34]. Some of these surfactants show an interesting molecular
aggregation behavior with the formation of twisted structures [40].
J.
Synthesis of Chemo-Cleavable Gemini Surfactants
For practical purposes, hydrolyzable cationic gemini surfactants were synthesized as
shown in Scheme 11 [2,41–43]. This possibility is an important one because some types
of cationic gemini surfactant may be hardly decomposed in the environment [42]. In
Scheme 11, when the lipophilic chain contains an ester linkage, it is preferable to use
quaternizing and spacer-forming reagents with higher reactivity, such as allylic and
propargylic dihalide [43], as satis
factory yield of the end product was not obtained with epichlorohydrin as the reagent
[42].
K.
Synthesis of Cationic Oligomeric Surfactants
Many types of oligomeric surfactant were synthesized. Indeed, their properties are
expected to be intermediate between those of monomeric and gemini surfactants, on the
one hand, and of polymeric surfactants, on the other hand. Oligomeric surfactants may
also show very special and specific properties.
Cationic oligomeric surfactants were prepared by using reactions 1 and 2 (Scheme 1),
with an appropriate synthetic procedure and starting compounds. Scheme 12 shows the
synthesis of a trimeric surfactants from monoammonium dibromide by reaction 1 [44]
and from N, N, N′, N″ ,N″-penta-methyldiethylenetriamine by reaction 2 [25]. A
tetrameric surfactant was
III.
ANIONIC GEMINI SURFACTANTS
A.
Synthesis of Typical Anionic Gemini Surfactants
Compounds having two hydrophilic groups and two lipophilic chains (i.e., gemini
surfactants) have been prepared since before 1988. However, the first report which
showed the properties of gemini surfactants as universal in
Gemini surfactants 20
SCHEME 19 Benzenesulfonate
gemini surfactants as components of
industrial product. (From Refs. 60 and
61.)
Synthesis of gemini (dimeric) and related surfactants 23
B.
Synthesis of Chemo-Cleavable Anionic Gemini Surfactants
By utilizing the ozone-cleavable character of carbon-carbon double bonds, a series of
gemini surfactants, the surface activity of which can be tuned at the user’s will, were
prepared as shown in Scheme 22 [65,66]. Acid-susceptible acetal-type anionic gemini
surfactants were prepared from diethyl tartrate according to Scheme 23 [67].
Gemini surfactants 24
SCHEME 23 Chemo-cleavable
gemini surfactants with tartrate moiety.
(From Ref. 67.)
Synthesis of gemini (dimeric) and related surfactants 25
C.
Synthesis of Multi-Armed Anionic Gemini Surfactants
Sulfonate, sulfate, and carboxylate gemini surfactants bearing an extra lipophilic chain,
which may affect the surface and aggregation properties, were prepared as shown in
Scheme 24 from diglycidyl compounds with glyceryl monoether [68] and diethanolamide
structures as the spacer group [69].
D.
Polyanionic Gemini and Oligomeric Surfactants
Trimethylolethane and pentaerythritol were utilized as starting compounds for triglycidyl
and tetraglycidyl ethers leading to sulfonate trimeric [70] (see Scheme 25) and sulfate
tetrameric [71] surfactants, respectively. Pentaerythritol also permitted the synthesis of a
polyanionic gemini surfactant with four sulfate groups, as shown in Scheme 25 [71].
IV.
NONIONIC GEMINI SURFACTANTS
Nonionic gemini surfactants are relatively newcomers in the field of gemini surfactants.
In view of the importance of conventional nonionic surfactants, they are likely to be
further investigated in the future.
Nonionic gemini surfactants having an amide structure and polyoxyethylene head
groups have been synthesized (Scheme 26). The reported results showed that the dimeric
structure provided a much lower cmc value than for the corresponding conventional
surfactant, as for ionic gemini surfactants [72].
Concerning carbohydrate-based nonionic gemini surfactants, sugar-amide and sugar-
amine gemini surfactants showing interesting molecular aggregation character were
prepared by direct reaction of D-glucose and α,ω-diaminoalkane under the hydrogenation
conditions, followed by acylation with carboxylic acid anhydride (see Scheme 27)
[73,74]. Saccharide lactones were also utilized via amidation as starting material for
constructing hydrophilic groups (Scheme 27) [39]. Because this kind of gemini and a
sugar-amine gemini mentioned here contain tertiary amino groups, they may show
cationic character.
Synthesis of gemini (dimeric) and related surfactants 27
SCHEME 26 Polyoxyethylene-type
nonionic gemini surfactants. (From
Ref. 72.)
V.
ZWITTERIONIC GEMINI SURFACTANTS
SCHEME 30 Bis(taurine)-type
anionic gemini surfactants from
diglycidyl ethers. (From Ref. 79.)
SCHEME 31 Complexane-type
gemini and trimeric surfactants. (From
Ref. 80.)
VI.
CHEMO-ENZYMATIC SYNTHESIS OF GEMINI SURFACTANTS
Combination of enzymatic and chemical processes can be effective for functional group-
selective or regioselective preparation of amino acid derivatives. Enzymatic process
using lipase was adopted for the preparation of both cationic and anionic gemini
surfactants in the step of esterification and amidation (see Scheme 33) [84,85].
Synthesis of gemini (dimeric) and related surfactants 31
VII.
CATANIONIC GEMINI SURFACTANTS
A catanionic ion pair, which is composed of one dicarboxylate species and two
amphiphilic sec-ammonium species bearing a sugar hydrophilic group, may be classified
as a member of gemini surfactants. Catanionic gemini surfactants shown in Scheme 34,
which were easily prepared in two steps from unprotected lactose, showed anti-HIV
(human immunodeficiency virus) activity and interesting molecular aggregation behavior
as well as surface activity [86].
SCHEME 33 Chemo-enzymatic
synthesis of gemini surfactants from
amino acids. (From Refs. 84 and 85.)
Gemini surfactants 32
VIII.
CONCLUSIONS
Currently, many reports concerning gemini surfactants are published every year. The
chemistry of gemini surfactants has developed considerably in the past 15 years, leading
to a diversification of the available chemical structures. In turn, this has led to a fuller
disclosure of their higher efficiency and/or novel properties and suggested the possibility
of new uses (see Chapter 13). In view of the importance of the function exerted by
surfactants in various heterogeneous systems, the development of gemini surfactant
chemistry should influence many fields, such as the chemical industry, agricultural
chemicals, pharmaceuticals, medical science, and so forth, although the evaluation of
gemini surfactants for practical uses is still under way. Further development of chemistry
and technology of gemini surfactants is expected.
REFERENCES
1. Bunton, C.A.; Robinson, L.; Schaak, J.; Stam, M.F. J. Org. Chem. 1971, 36, 2346–2350.
2. Devinsky, F.; Masarova, L.; Lacko, I. J. Colloid Interf. Sci. 1985, 105, 235–239.
3. Devinsky, F.; Lacko, I.; Bittererova, F.; Tomeckova, L. J. Colloid Interf. Sci. 1986, 114, 314–
322.
4. Rosen, M.J. CHEMTECH 1993, 23, 30–33.
5. Nakatsuji, Y.; Ikeda, I. Chim. Oggi 1997, 40–43.
6. Zana R. In Novel Surfactants. Preparation, Applications and Biodegradability; Surfactant
Science Series; Holmberg, K., Ed.; Marcel Dekker: New York, 1998; 241–277.
Synthesis of gemini (dimeric) and related surfactants 33
7. Menger, F.M.; Keiper, J.S. Angew. Chem. Int. Ed. 2000, 39, 1906–1920.
8. In, M. In Gemini Surfactants and Surfactant Oligomers. Reactions and Synthesis in Surfactant
Systems; Surfactant Science Series; Texter, J., Ed.; Marcel Dekker: New York, 2001; 59–110.
9. Zana, R.; Benrraou, M.; Rueff, R. Langmuir 1991, 7, 1072–1075.
10. Kim, T.S.; Hirao, T.; Ikeda, I. J. Am. Oil Chem. Soc. 1996, 73, 67–71.
11. Dreja, M.; Gramberg, S.; Tieke, B. Chem. Commun. 1998, 1371–1372.
12. Dreja, M.; Tieke, B. Ber. Bunsenges. Phys. Chem. 1998, 102, 1705–1709.
13. De, S.; Aswal, V.K.; Goyal, P.S.; Bhattacharya, S. J. Phys. Chem. B 1998, 102, 6152–6160.
14. Rosen, M.J.; Liu, L. J. Am. Oil Chem. Soc. 1996, 73, 885–890.
15. Menger, F.M.; Littau, C.A. J. Am. Chem. Soc. 1991, 113, 1451–1452.
16. Menger, F.M.; Littau, C.A. J. Am. Chem. Soc. 1993, 115, 10083–10090.
17. Hattori, N.; Yoshino, A.; Okabayashi, H.; O’Connor, C.J. J. Phys. Chem. B 1998, 102, 8965–
8973.
18. Tatsumi, T.; Zhang, W.; Nakatsuji, Y.; Miyake, K.; Matsushima, K.; Tanaka, M.; Furuta, T.;
Ikeda, I. J. Surfact. Detergents 2001, 4, 271–277.
19. Menger, F.M.; Keiper, J.S.; Azov, V. Langmuir 2000, 16, 2062–2067.
20. Jenkins, K.M.; Wettig, S.D.; Verrall, R.E. J. Colloid Interf. Sci. 2002, 247, 456–462.
21. Brinchi, L.; Germani, R.; Goracci, L.; Savelli, G.; Bunton, C.A. Langmuir 2002, 18, 7821–
7825.
22. Buwalda, R.T.; Engberts, J.B.F.N. Langmuir 2001, 17, 1054–1059.
23. Stathatos, E.; Lianos, P.; Rakotoaly, R.H.; Laschewsky, A.; Zana, R. J. Colloid Interf. Sci.
2000, 227, 476–481.
24. Li, F.; Rosen, M.J. J. Colloid Interf. Sci. 2000, 224, 265–271.
25. Esumi, K.; Taguma, K.; Koide, Y. Langmuir 1996, 12, 4039–4041.
26. Liu, L.; Rosen, M.J. J. Colloid Interf. Sci. 1996, 179, 454–459.
27. Li, Z.X.; Dong, C.C.; Thomas, R.K. Langmuir 1999, 15, 4392–4396.
28. Ricci, C.G.; Cabrera, M.I.; Luna, J.A.; Grau, R.J. Synlett2002, 1811–1814.
29. Zhu, Y.-P.; Ishihara, K.; Masuyama, A.; Nakatsuji, Y.; Okahara, M. J. Jpn. Oil Chem. Soc.
1993, 42, 161–167.
30. Oda, R.; Huc, I.; Candau, S. J. Chem. Commun. 1997, 2105–2106.
31. Oda, R.; Huc, I.; Danino, D.; Talmon, Y. Langmuir 2000, 16, 9759–9769.
32. Cerichelli, G.; Luchetti, L.; Mancini, G.; Savelli, G. Langmuir 1999, 15, 2631–2634.
33. Perez, L.; Torres, J.L.; Manresa, A.; Solans, C.; Infante, M.R. Langmuir 1996, 12, 5296–5301.
34. McGregor, C.; Perrin, C.; Monck, M.; Camilleri, P.; Kirby, A.J. J. Am. Chem. Soc. 2001, 123,
6215–6220.
35. Camilleri, P.; Kremer, A.; Edwards, A.J.; Jennings, K.H.; Jenkins, O.; Marshall, I.; McGregor,
C.; Neville, W.; Rice, S.Q.; Smith, R.J.; Wilkinson, M.J.; Kirby, A. J. Chem. Commun. 2000,
1253–1254.
36. Ronsin, G.; Perrin, C.; Guedat, P.; Kremer, A.; Camilleri, P.; Kirby, A. J. Chem. Commun.
2001, 2234–2235.
37. Ronsin, G.; Kirby, A.J.; Rittenhouse, S.; Woodnutt, G.; Camilleri, P. J. Chem. Soc. Perkin.
Trans.2002, 2, 1302–1306.
38. Menger, F.M.; Mbadugha, B.N.A. J. Am. Chem. Soc. 2001, 123, 875–885.
39. Wilk, K.A.; Syper, L.; Domagalska, B.W.; Komorek, U.; Maliszewska, I.; Gancarz, R. J.
Surfact. Detergents 2002, 5, 235–244.
40. Oda, R.; Huc, I.; Candau, S.J. Angew. Chem. Int. Ed.1998, 37, 2689–2691.
41. Kim, T.S.; Tatsumi, T.; Kida, T.; Nakatsuji, Y.; Ikeda, I. J. Jpn. Oil Chem. Soc. 1997, 46, 747–
753.
42. Tatsumi, T.; Zhang, W.; Kida, T.; Nakatsuji, Y.; Ono, D.; Takeda, T.; Ikeda, I. J. Surfact.
Detergents 2000, 3, 167–172.
43. Tatsumi, T.; Zhang, W.; Kida, T.; Nakatsuji, Y.; Ono, D.; Takeda, T.; Ikeda, I. J. Surfact.
Detergents2001, 4, 279–285.
Gemini surfactants 34
44. Zana, R.; Levy, H.; Papoutsi, D.; Beinert, G. Langmuir 1995, 11, 3694–3698.
45. Zana, R.; In, M.; Levy, H. Langmuir 1997, 13, 5552–5557.
46. Kim, T.S.; Kida, T.; Nakatsuji, Y.; Ikeda, I. Langmuir 1996, 12, 6304–6308.
47. Menger, F.M.; Migulin, V.A. J. Org. Chem. 1999, 64, 8916–8921.
48. Okahara, M.; Masuyama, A.; Sumida, Y.; Zhu, Y.-P. J. Jpn. Oil Chem. Soc. 1988, 37, 746–748.
49. Zhu, Y.-P.; Masuyama, A.; Okahara, M. J. Am. Oil Chem. Soc. 1990, 67, 459–463.
50. Zhu, Y.-P.; Masuyama, A.; Nagata, T.; Okahara, M. J. Jpn. Oil Chem. Soc. 1991, 40, 473–477.
51. Zhu, Y.-P.; Masuyama, A.; Nakatsuji, Y.; Okahara, M. J. Jpn. Oil Chem. Soc. 1993, 42, 86–94.
52. Zhu, Y.-P.; Masuyama, A.; Okahara, M. J. Am. Oil Chem. Soc. 1991, 68, 268–271.
53. Zhu, Y.-P.; Masuyama, A.; Kobata, Y.; Nakatsuji, Y.; Okahara, M.; Rosen, M.J. J. Colloid
Interf. Sci. 1993, 158, 40–45.
54. Zana, R.; Levy, H.; Kwetkat, K. Langmuir 1997, 13, 402–408.
55. Alami, E.; Holmberg, K. J. Colloid Interf. Sci. 2001, 239, 230–240.
56. Renouf, P.; Mioskowski, C.; Lebeau, L. Tetrahedron Lett. 1998, 39, 1357–1360.
57. Van Zon, A.; Bouman, J.T.; Deuling, H.H.; Karaborni, S.; Karthaeuser, J.; Mensen, H.T.G.A.;
Van Ost, N.M.; Raney, K.H. Tenside Surfact. Detergents 1999, 36, 84–86.
58. Jpn. Tokkyo Kokai Koho 1995, Japanese Patent to Lion Corporation. 7–11289.
59. Jpn. Tokkyo Kokai Koho 1994, Japanese Patent to Lion Corporation. 6–172784. 1994, 6–
65592. 1994, 6–330084.
60. Quencer, L.B.; Kokke-Hall, S.; Loughney, T. In Proceedings of the 4th World Surfactants
Congress 1996; Vol. 2.66–75.
61. Rosen, M.J.; Li, F. J. Colloid Interf. Sci. 2001, 234, 418–424.
62. Duivenvoorde, F.L.; Feiters, M.C.; van der Gaast, S.J.; Engberts, J.B.F.N. Langmuir 1997, 13,
3737–3743.
63. Sommerdijk, N.A.J.M.; Hoeks, T.H.L.; Synak, M.; Feiters, M.C.; Nolte, R.J.M.; Zwanenburg,
B. J. Am. Chem. Soc. 1997, 119, 4338–4344.
64. Jpn. Tokkyo Kokai Koho 2000, Japanese Patent to Chukyo-Yushi Co. Ltd. 12–219654.
65. Masuyama, A.; Endo, C.; Takeda, S.; Nojima, M. Chem. Commun., 2023–2024 (1998).
66. Masuyama, A.; Endo, C.; Takeda, S.; Nojima, M.; Ono, D.; Takeda, T. Langmuir 2000, 16,
368–373.
67. Ono, D.; Tanaka, T.; Masuyama, A.; Nakatsuji, Y.; Okahara, M. J. Jpn. Oil Chem. Soc. 1993,
42, 10–16.
68. Zhu, Y.-P.; Masuyama, A.; Kirito, Y.; Okahara, M. J. Am. Oil Chem. Soc. 1991, 68, 539–543.
69. Zhu, Y.-P.; Masuyama, A.; Kirito, Y.; Okahara, M.; Rosen, M.J. J. Am. Oil Chem. Soc. 1992,
69, 626–632.
70. Masuyama, A.; Yokota, M.; Zhu, Y.-P.; Kida, T.; Nakatsuji, Y. J. Chem. Soc. Chem. Commun.
1994, 1435–1436.
71. Murguia, M.C.; Grau, R.J. Synlett 2001, 1229–1232.
72. Paddon-Jones, G.; Regismond, S.; Kwetkat, K.; Zana, R. J. Colloid Interf. Sci. 2001, 243, 496–
502.
73. Pestman, J.M.; Terpstra, K.R.; Stuart, M.C.A.; van Doren, H.A.; Brisson, A.; Kellogg, R.M.;
Engberts, J.B.F.N. Langmuir 1997, 13, 6857–6860.
74. Fielden, M.L.; Perrin, C.; Kremer, A.; Bergsma, M.; Stuart, M.C.; Camilleri, P.; Engberts,
J.B.F.N. Eur. J. Biochem. 2001, 268, 1269–1279.
75. Gao, C.; Millqvist-Fureby, A.; Whitcombe, M.J.; Vulfson, E.N. J. Surfact. Detergents 1999, 2,
293–302.
76. Castro, M.J.L.; Kovensky, J.; Cirelli, A.F. Tetrahedron Lett. 1997, 38, 3995–3998.
77. Castro, M.J.L.; Kovensky, J.; Cirelli, A.F. Tetrahedron 1999, 55, 12,711–12,722.
78. Castro, M.J.L.; Kovensky, J.; Cirelli, A.F. Langmuir 2002, 18, 2477–2482.
79. Masuyama, A.; Hirono, T.; Zhu, Y.-P.; Okahara, M.; Rosen, M.J. J. Jpn. Oil Chem. Soc. 1992,
41, 301–305.
80. Onitsuka, E.; Yoshikura, T.; Koide, Y.; Shosenji, H.; Esumi, K. J. Oleo Sci. 2001, 50, 159–164.
Synthesis of gemini (dimeric) and related surfactants 35
I.
INTRODUCTION
II.
MODELS OF SURFACTANT SELF-ASSEMBLY
A.
Surface Behavior
Let us start by considering an aqueous surfactant solution below the cmc. The soluble
surfactant molecules self-assemble into a condensed layer at the water-air interface,
referred to as a Gibbs monolayer (to be distinguished from a Langmuir monolayer that
forms when insoluble surfactants are spread on the water-air interface) [22]. Because the
surfactant is water soluble, this layer exchanges molecules with the bulk solution and a
nonuniform concentration profile forms. Typical surfactants have strong surface activity
(i.e., the energy gained by a molecule when it migrates to the surface is much larger than
Models of gemini surfactants 39
the thermal energy kBT). As a result, the concentration profile drops sharply to its bulk
value within a molecular distance from the surface (hence the term monolayer).
The number of molecules participating in a Gibbs monolayer per unit area, the surface
excess Γ, is obtained by integrating the excess concentration (with respect to the bulk
one) over the entire solution. Such a monolayer can be regarded as a separate subsystem
at thermodynamic equilibrium and in contact with a large reservoir of molecules having
temperature T and chemical potential µ. From the excess free energy per unit area of this
system, γ(T, µ), which is by definition the surface tension of the solution, we get the
number of molecules per unit area:
(1)
This is referred to as the Gibbs equation [22]. For dilute solutions µ α kBT In C, where C
is the bulk surfactant concentration. (The constant of proportionality is 1 for nonionic
surfactants and ionic surfactants at a high salt concentration; it has a higher value for salt-
free ionic surfactant solutions, where strong correlations between the different ions lead
to nonideal activity coefficients [22].) Hence,
(2)
(3)
Gemini surfactants 40
This is merely a harmonic approximation for the interaction energy associated with small
deviations from optimal packing.
We can slightly modify this result to obtain a similar harmonic estimate for the energy
u2 of effective interaction between two neighboring molecules residing in the saturated
monolayer. (This will be useful later when we add the spacer to form gemini surfactants.)
We need to divide u by half the number of in-plane neighbors, q/2, and express a in terms
of the average intermolecular distance r, a=ηr2, where η is a prefactor of order unity
(e.g., for hexagonal packing q=6 and Assuming, again, that r is close to
its optimal value r0, we get
(4)
B.
Micelles
As the solution concentration is increased beyond the cmc, the surfactant molecules start
to form aggregates. Unlike simple solute molecules (e.g., alkanes), which undergo
macroscopic phase separation upon increasing concentration or changing temperature,
surfactants form micelles at the mesoscopic scale. The challenges posed to theories of
surfactant self-assembly are to predict the micellization point as a function of
concentration (i.e., the cmc, hereafter referred to by the corresponding volume fraction
φcmc) and temperature (Tm), as well as the micelle shape and size. The main
complications come from the fact that micellization is not a macroscopic phase
transition—the aggregate sizes are finite and polydisperse—and, thus, the well-developed
theoretical framework of phase transitions does not strictly apply.
From a thermodynamic point of view, the difference between surfactant micellization
and phase separation lies in the following observation [12]. For alkanes solubilized in
water, for example, the (Gibbs) free energy per molecule in an aggregate of size N, gN, is
a monotonously decreasing function of N—for N→∞, gN tends to the free energy per
molecule in the bulk alkane phase, g∞, whereas for smaller N, gN>g∞ due to unfavorable
surface terms of the finite cluster. As a result, there is a critical concentration (or critical
temperature) at which the favorable size changes discontinuously from monomers
solubilized in water (N = 1) to a macroscopic phase of bulk alkane (N → ∞). The first-
order phase transition point is reached when the chemical potential of monomers exceeds
min {gN} = g∞. In a dilute solution, this implies that φc=exp[g∞/(kBT)], kBTc=g∞/ln φ.
(We have set the free energy of the N =1 state as the reference, g1=0.) In the case of
surfactants, by contrast, gN has a minimum at a finite aggregate size N*. As a result, when
Models of gemini surfactants 41
the chemical potential exceeds gN*, a large population of aggregates appears, whose sizes
are distributed around N*. Hence, the micellization point can be estimated as
(5)
The remaining task is to obtain a theoretical expression for the free energy per
molecule in aggregates of size N and shape S. Because we expect this function to have a
minimum at a finite yet large N (say, N* ~ 102), the importance of many-body
interactions is inevitable, and obtaining from rigorous statistical mechanics is a
formidable task. Consequently, analytical models have relied on phenomenological
approaches, trying to account for various competing contributions to the free energy
while assuming a certain geometry for the aggregate [11,12,23]. From the minimum of
with respect to N and various possible shapes S, one can obtain the aggregate shape,
aggregation number, and micellization point, using Eq. (5).
In the simplest picture [11,12], rough estimates for the minimum of can be
obtained by imposing geometrical constraints that arise from the incompressibility of the
micellar hydrocarbon core (see Fig. 1). These constraints lead to a finite aggregation
number N*, as required. The hydrocarbon tail chains cannot extend beyond a certain
length l, and each tail must occupy a certain volume v. (Both l and v are known to have a
simple linear dependence on the number of hydrocarbon groups in the tail chain [24].) In
addition, as in the case of the Gibbs monolayer of Section II.A, surfactant molecules in
the favorable aggregate state attain a certain optimal area per molecule a0. Then, due to
translational entropy, the favorable aggregates would be the smallest
Gemini surfactants 42
(6)
If P<⅓, the constraints can be satisfied by spherical micelles, which will be the smallest
and hence the most favorable ones; when ⅓P<½, the micelles will have to be of
elongated or cylindrical shapes; for ½<P<1, planar shapes will form; and for P>1, the
morphology must be inverted.
Once the shape is determined, we can find the maximum allowed aggregation number.
For example, for spherical micelles N<4πl2/a0=(4π/3)l3/v, and we get
(7)
We see how competing interactions between the molecules (giving rise to a0) together
with the incompressibility of the micellar core lead to finite micelles. Because the tail
chains usually should not stretch to their full extent, the actual aggregation number will
be smaller than this upper bound.
Yet, these geometrical arguments cannot provide us with theoretical predictions as to
the optimal molecular area a0 itself or the aggregation free energy , as well as their
dependence on parameters such as temperature or salt concentration. In order to get such
information and subsequently predict the micellization point, micelle shape, and size, one
needs a more detailed theory.
1.
Phenomenological Models
There have been attempts to analytically account for the various competing contributions
to the free energy per molecule (e.g., Ref. 23). The advantage of this approach is that
once we have an expression for the free energy, we can easily change parameters and
gain insight into the role of various contributions. On the other hand, such models
essentially attempt to push the limits of the phenomenological approach toward a detailed
molecular description. They usually entail uncontrolled approximations and parameters
Models of gemini surfactants 43
whose accurate values are often hard to obtain. As an example, which will serve us in
Section III, we give an analysis along the lines of (yet not identical to) Ref. 23.
Five major contributions to the aggregation free energy (per surfactant molecule on
the aggregate) can be considered [23]:
(8)
1. The driving force for aggregation is the hydrophobic effect (i.e., the free energy per
surfactant molecule ghc gained by shielding the hydrocarbon groups from water) [24].
This contribution to is negative and, to a good approximation, independent of N
and the aggregate geometry S; namely its contribution to the entire aggregate free
energy is linear in N and tends to increase the aggregate size. The hydrophobic term
ghc depends linearly on the number of hydrocarbon groups in the surfactant, with a
reduction of roughly kBT per hydrocarbon group [12]. That is why, for regular
surfactants, the cmc decreases exponentially with the number of hydrocarbon groups
in the molecule and is reduced by a factor of roughly 2–3 per each additional
hydrocarbon group.
2. The hydrophobic gain is corrected by an interfacial contribution gint due to the
unfavorable contact between the hydrocarbon core and water:
(9)
where γ1 is the interfacial tension of the core-water interface (roughly equal to the
hydrocarbon-water interfacial tension) and amin is the minimum area per molecule
(i.e., the interfacial area occupied by a head group). This contribution evidently
acts to reduce the area per molecule.
3. If the surfactant head groups are charged, there is electrostatic repulsion between them,
acting to increase a. Within the Poisson-Boltzmann theory, this electrostatic
contribution is given by [25]
(10)
Gemini surfactants 44
(11)
5. The last contribution to the free energy is associated with the tail packing in the
hydrophobic core; that is, deviations of the hydrocarbon tail chains from their relaxed
length l0,
(12)
The elastic constant k′ depends on the chain statistics, as well as the packing
parameter (i.e., aggregate shape) [23].
The equilibrium aggregation number N and specific area a (and hence also
aggregate size R) for a given shape S and surfactant chemical potential µ are then
determined by the equations
(13)
Comparing the minimum value of for various shapes S, one also obtains the
equilibrium aggregate morphology. As long as µ<kBT In φcmc, these equations will have
no solution, and the monomeric state (N =1) of single surfactant molecules solubilized in
water is the stable one. As the chemical potential increases, we reach the micellization
condition given by Eq. (5), where the average micelle size at the cmc, N*, can be
2.
Computer Simulations
Another route to overcome the complexity of treating surfactant micellization is to use
computer simulations. This approach can be divided into two categories: statistical-
mechanical models using Monte Carlo (MC) simulations and molecular dynamics (MD)
simulations.
Following Widom’s statistical-mechanical model of microemulsions [26], a host of
lattice models was presented for treating surfactant self-assembly (see, e.g., Refs. 27–31).
Models of gemini surfactants 45
These molecular toy models represent the water molecules and various groups in the
surfactant as Ising spins on a discrete lattice. The various interactions between the groups
are represented by ferromagnetic or antiferromagnetic couplings between nearest-
neighbor spins (see Fig. 2). Evidently, this is a very crude description of surfactant
solutions and is not expected to yield quantitative predictions. Another difficulty is
attaining thermodynamic equilibrium in simulations of these self-assembling
(14)
where J>0 is the coupling strength, (ij) denotes summation only over nearest-neighbor
pairs of the lattice, and δi,j=1 when i=j and zero when i≠j is the Kronecker delta function.
Two neighboring water molecules attract each other with energy −J because, then, σiσj=1.
The same applies to two tail groups, whereas a water molecule and a tail group repel with
energy +J. The extra factor of (1—2δσi,2δσi,2) in Eq. (14) is unity for all cases except
when i and j are two heads with σ=+2, yielding a repulsion of +4J between two head
groups. Finally, a head group and a water molecule attract with energy −2J, and a head
group and a tail group repel with energy+2J. These couplings apply whether the two
neighboring lattice sites belong to the same molecule or not. In addition, the groups
belonging to the same surfactant molecule are kept linked throughout the simulation.
Thus, the essential features of hydrophobicity, hydrophilicity, molecular connectivity,
and (screened) electrostatic repulsion are all accounted for using the single parameter J.
Other parameters are the length of the tail group and total number of surfactant molecules
in the system.
The MC simulation starts from a certain configuration of surfactant molecules in
water. At each iteration, the various groups of the surfactants are moved while
maintaining the connectivity of the molecules, their total number, and the total number of
water molecules (e.g., using a “slithering snake” scheme [31]). The energetic cost of the
move is calculated using Eq. (14), and the MC step is accepted or rejected according to a
Metropolis criterion, ensuring convergence toward equilibrium.
This simple scheme can reproduce much of the richness of surfactant self-assembly,
including the formation of monolayers, micelles, and bilayers, the dependence of the cmc
on tail length, transitions between various aggregate shapes, and so forth. On the other
Models of gemini surfactants 47
hand, such models can merely indicate general trends and not detailed information. For
example, the correspondence between the MC spin variable representing subgroups of a
surfactant molecule and the actual chemical groups is not well defined and remains
ambiguous to some extent.
Another class of numerical studies that have been used to explore surfactant self-
assembly are molecular dynamics (MD) simulations. These models range in detail from
coarse-grained bead-spring representations of the molecules (e.g., Refs. 32–37) to
atomistic descriptions (e.g., Refs. 38–42). The advantage of the MD approach, as
compared to phenomenological theories and spin models, is that the description of the
system on the molecular scale is less artificial. The disadvantages are the limited spatial
and temporal extent of the simulations, entailing equilibration problems, and sometimes
also a large number of required parameters. A typical all-atom MD simulation of an
aqueous surfactant system may contain about 100 surfactant molecules along with a few
thousands of water molecules, and the dynamics can be run for a few nanoseconds (e.g.,
Ref. 42). A coarse-grained simulation allows a significant increase of these numbers at
the expense of molecular detail (see, e.g., Ref. 37). Here, we outline a coarse-grained
approach to surfactant micellization, as presented in Ref. 32, which was later extended to
gemini surfactants [36].
The MD model of Ref. 32 contains only two types of particles: waterlike and oillike,
where a surfactant molecule is composed of a few waterlike particles (the head group)
and a chain of oillike particles (the tail). The particles interact via a truncated Lennard-
Jones potential,
(15)
C.
Phase Behavior
Concentrated surfactant solutions and ternary water-oil-surfactant systems exhibit a rich
variety of disordered and liquid-crystalline phases [43–45]. Some examples are the
lamellar (Lα) phase, sponge (L3) phase, hexagonal (H1) phase, and cubic (V1) phase. All of
these phases are based on various packing of surfactant layers: bilayers (in the binary-
mixture case) or monolayers (in the ternary case). The lamellar phase is made of stacks of
parallel layers, the sponge phase contains a disordered arrangement of multiconnected
layers, the hexagonal phase consists of hexagonal arrays of parallel cylinders, and the
cubic phase contains spherical layers arranged in a cubic lattice.
Unlike micellization, one deals here with macroscopic bulk phases and their
corresponding phase transitions. Hence, the powerful tools of thermodynamics and
statistical mechanics are applicable. Consequently, the theory of surfactant phase
behavior has reached a more advanced level, particularly in the case of phases with long-
ranged order. We will not review these theories here, as most of them have not been used
in current models of gemini surfactants, but we will merely mention the various
approaches.
Two phenomenological approaches to the phase behavior of surfactant binary and
ternary mixtures have been used. The first is based on the Ginzburg-Landau formalism,
which has been widely used in statistical physics [43]. It starts with a lattice description
of the mixture and derives from it a coarse-grained, continuous expression for the energy
(Hamiltonian), which can be studied by statistical-mechanical techniques. The second
approach is based on the elastic and thermodynamic properties of the membranes that
make the various phases. For a review, see, for example, Ref. 45. In addition to these
phenomenological theories, a variety of lattice spin models employing Monte Carlo
simulations, as discussed in Section II.B, were originally designed and applied to study
surfactant phase behavior [26–31].
III.
MODELS OF GEMINI SURFACTANTS
Having provided the necessary background, we now turn to models of gemini surfactants.
As will be demonstrated in this section, these models are essentially extensions of
surfactant self-assembly theories, which have been reviewed in Section II. The binding of
the surfactant molecules into pairs via spacer chains introduces new constraints affecting
the molecular arrangement in monolayers, micelles, and mesophases, as well as the
thermodynamics of self-assembly.
A.
Surface Behavior
As in Section II, we begin by looking at a saturated monolayer, this time made of gemini
surfactants, lying at the water-air interface. The gemini nature of the molecules (i.e., the
introduction of the spacer) adds considerable complexity to the problem, mainly because
it introduces anisotropy and inhomogeneity into the monolayer. A schematic view of the
Models of gemini surfactants 49
surface covered with dimers is shown in Fig. 3. The dimers may be oriented in various
directions, and the distances between two linked monomers (in a dimer) and between two
unlinked ones will differ in general.
Nevertheless, we are going to disregard these complications and focus on the simplest
question: How does the introduction of a spacer consisting of s groups affect the in-plane
distance rd(s) between two monomers belonging to the same dimer? Although seemingly
oversimplified, the answer to this question will give us key insight into the surface
behavior of gemini surfactants.
We proceed by reviewing and slightly extending the work presented in Refs. 46 and
47. We have seen in Section II.A that the interaction between surfactants in a saturated
monolayer can be roughly approximated by effective springs whose equilibrium length r0
is determined by the optimum packing at saturation and whose spring constant k0 is given
in Eq. (4). Thus,
(16)
The origin of the spacer spring is entropy and its parameters are determined by the
statistical distribution of spacer configurations. The equilibrium length of the spring is the
mean end-to-end distance of the spacer chain, and the spring constant is inversely
proportional to the variance of the endto-end distance,
(17)
where the averages are taken over all spacer chain configurations. Thus, the harmonic
spring approximation for the spacer is equivalent to representing the actual statistical
distribution of spacer configurations by its first two moments.
Before considering specific models for the spacer chain, let us examine what
qualitative results are expected from this description. When the spacer is very short and
rigid, such that , the equilibrium length rd of the dimer is determined by the
spacer, rd rs. On the other hand, when the spacer is very long and flexible, such that
, rd will be determined by the regular monomer-monomer interaction, rd r0.
Hence, upon increasing the number s of groups in the spacer, we expect rd(s) to first
increase and then saturate toward r0, the optimal distance between the monomeric
surfactants. Whether the behavior for intermediate spacer lengths is monotonous or not
depends on specific details of the spacer chain. If the spacer stiffness ks(s) drops
sufficiently fast with s, the interaction spring will start dominating before rs(s) exceeds r0,
and rd(s) will then grow monotonously with s. By contrast, if ks(s) decreases slowly with
s, the spacer spring will dominate even for quite long spacers and rd rs will become
larger than r0. For even longer s, it will have to decrease back toward r0, leading to
nonmonotonous behavior in this case.
The simplest model for the spacer is that of a Gaussian, constraint-free chain. This
case is somewhat artificial and is discussed here merely as a model for very flexible and
long chains, in contrast with more realistic models discussed in the following for more
rigid chains. A Gaussian chain consisting of s segments is analogous to a random walk of
s steps. The mean squared displacement of such a walk, averaged over all s-step walks,
should scale linearly with s. The mean end-to-end distance of a Gaussian spacer is
therefore rs~bs½, where b is the segment length. More specifically, the statistical
distribution of the end-to-end distance in a Gaussian chain is
(18)
From the mean and variance of this distribution, we get, according to Eq. (17),
(19)
Thus, in order to calculate rs and ks, we just need to know the segment length b. For a
polymethylene spacer, b is 2.53Å. The remaining information required to compute rd
Models of gemini surfactants 51
from Eq. (17) are the properties of the monomeric surfactant in a saturated monolayer,
namely k0 and r0. In a roughly hexagonal arrangement of molecules, one has
(see Section II.A). A saturated monolayer of dodecyltrimethylammonium
bromide (DTAB) surfactants, for example, is known to have .
By using these values and substituting Eq. (19) in Eq. (16), rd as function of s is
calculated and depicted as the dashed line in Fig. 4. The intermonomer distance increases
moderately with s and even exceeds r0, yet the maximum and descent back to r0 are
shallow and occur at very large s, lying outside the experimentally relevant range of
spacer lengths 1≤s≤20 shown in Fig. 4.
More realistically, the spacer chain can be described by the rotationalisomeric model,
where each segment in the chain can have only three possible orientations with respect to
the two that precede it in the sequence (the three conformations called trans and gauche±)
[48]. In addition, we require that the hydrophobic spacer be restricted to reside in the
nonaqueous side of the water-air interface. These constraints stiffen the chain and bias its
statistics toward larger end-to-end distances. Therefore, we expect a larger overshoot and
a sharper maximum of rd(s), as is confirmed by the solid curve in Fig. 4. The points of
this curve were obtained from simulations of rotationalisomeric polymethylene chains
whose ends were fixed to a surface (the airwater interface) and whose segments were
forbidden from crossing that surface into the water side (see Ref. 46 for more details).
From the simulations, one obtains the end-to-end distance distribution and then extracts
the spring parameters rs and ks according to Eq. (17) [47].
What has been calculated is the intermonomer distance in a dimer and not the average
area per dimer in the monolayer. However, because the latter
presence of the tails is important for quantitative predictions, it is not expected to alter the
above-described qualitative competition picture, and a similar nonmonotonous behavior
of a(s) was indeed observed in bolaform surfactants as well [49].
B.
Micelles
The micellization behavior of gemini surfactants is qualitatively different from that of
regular ones. We have reviewed some of these differences in Section I, and they can now
be further elucidated in the light of what we have discussed in the previous sections.
The cmc of gemini surfactants is typically one to two orders of magnitude lower than
that of the corresponding monomeric surfactants [6]. The lower cmc can be directly
attributed to the increase in the number of hydrocarbon groups in the molecule (i.e.,
decrease in the hydrophobic contribution ghc) due to the second tail and also due to the
hydrophobic spacer chain in the case of m-s-m surfactants. Based only on the contribution
of a second tail to ghc and the fact that the molecular volume is roughly doubled going
from the monomeric surfactant to a gemini one, one would have predicted a larger
decrease in cmc than what is actually observed. The difference is probably due to
unfavorable terms introduced by the spacer, which will be further discussed in this
subsection.
The cmc of m-s-m gemini surfactants, instead of monotonously decreasing with the
number s of spacer hydrocarbon groups (i.e., with molecular hydrophobicity), is a
nonmonotonous function with a maximum around s 4–6 [6–8]. A corresponding
nonmonotonous behavior is observed in the Krafft temperature [9] and micellization
enthalpy [10]. This behavior can be attributed to the straightness and rigidity of short
spacers, which force their hydrocarbon groups to be in unfavorable contact with water. At
about s 4–6, although the spacer chain is still rigid, a gauche conformation should
become accessible, allowing some of the groups to penetrate in the micellar hydrophobic
core. When the spacer is hydrophilic, this effect should be absent, as indeed is the case
with m-EOx-m surfactants, exhibiting a weak monotonous increase of the cmc with the
hydrophilic spacer length x [4].
As a function of spacer length s, m-s-m surfactants exhibit an unusual progression of
aggregate shapes from cylinders to spheres to bilayers. This is different from the more
natural succession, occurring in monomeric surfactants, where the change in aggregate
curvature is monotonous: spheres transforming into cylinders transforming into bilayers.
Assuming that the molecular area at the aggregate surface is related to that in a saturated
monolayer, this uncommon behavior can be qualitatively understood in view of the
nonmonotonous variation of a(s) as a function of s, as discussed in Section III.A.
Considering that the radius and volume of the micellar core depend primarily on the tails
and not on the spacers, an increase and then a decrease of a as a function of s should be
accompanied by a decrease and then an increase in the packing parameter P of Eq. (6),
hence the unusual morphological sequence.
More specific predictions require a detailed theory and will be reviewed next.
Gemini surfactants 54
1.
Phenomenological Model
An extension of the phenomenological theory of surfactant aggregation to gemini
surfactants with hydrophobic spacers is presented in Ref. 50. It introduces the following
additions and modifications to the model outlined in Section II.B.
1. The hydrophobic free energy ghc contains, apart from the double-tail contribution, a
spacer contribution also. Only the spacer section which penetrates in the micellar
hydrophobic core, Score, is considered. This spacer section is taken simply as the
difference between the total spacer length and the mean head-head distance, Score=s—
a½/b. Because the three chains (two tails and spacer) are already in partial contact prior
to aggregation, the hydrophobic energy gain per hydrocarbon group is taken to be
smaller than in the case of a single-tail surfactant.
2. The interfacial term, gint(a), is modified to account for the part of the core-water
interfacial area that is now occupied by the spacer. The chain length that participates
in this shielding is proportional to a½. This contribution is thus
(20)
where γ2 is the spacer-water interfacial tension and w is the spacer width. If the
spacer is a polymethylene chain, then γ2=γ1 and this correction vanishes.
3. When the spacer is short, it forces the two tails to be closer together than they would be
if they belonged to two separate molecules. This packing constraint reduces the
entropy of the tail chains. For a single-tail surfactant, the area close to the core-water
surface sampled by tail groups is atail ~ ν/R (cf. Fig. 1), with a prefactor of order unity
that varies with aggregate shape [23,50]. The proximity to a second tail due to the
spacer reduces this available area per tail to asp ~ (sb)2. Thus, the contribution to asp
the free energy can be estimated as
(21)
4. Finally, the most difficult modification to handle is the electrostatic one. A short spacer
forces the distance between two connected head groups to be shorter than that between
two unconnected ones, resulting in a nonuniform charge distribution of pairs over the
micellar surface (cf. Fig. 3). This problem is bypassed in Ref. 50 by introducing an
empirical correction factor to ges, which becomes equal to unity when the spacer is
longer than the mean interhead distance.
This extended phenomenological model is applied in Ref. 50 to gemini surfactants with
short hydrophobic spacers, using parameters known from regular single-tail and double-
tail surfactants. The model yields cmc values for various tail lengths in good agreement
with the measured ones. More important, it correctly accounts for the observed micelle
shapes of m-s-m surfactants with small s (i.e., the crossover from cylinders to spheres as s
Models of gemini surfactants 55
is increased). Following the changes in the various free-energy contributions, one can
identify the crossover mechanism as a competition between the elastic and packing
contributions from the tails (favoring cylinders), and the electrostatic contribution
(favoring spheres). Note, however, the various assumptions and approximations involved
in these calculations. Although the electrostatic contribution to the free energy is found to
be crucial for the self-assembly behavior, it is treated somewhat dubiously, as already
admitted in Ref. 50.
2.
Computer Simulations
The additional complexity introduced by the spacers makes analytical calculations very
difficult. One is inclined, therefore, to resort to computer simulations in order to gain
detailed information on the self-assembly of gemini surfactants.
The statistical-mechanical approach based on Monte Carlo (MC) simulations, as
outlined in Section II.B, was extended to treat gemini surfactants [51,52]. The spin
assignment to various groups and the corresponding energy function are the same as for
regular, monomeric surfactants [see Eq. (14)]. The main modification is the connection of
head groups in pairs via spacers (Fig. 5). Both hydrophobic spacers (spins σ=−1) and
hydrophilic ones (σ=+1) were simulated. In addition, the role of spacer stiffness was
checked by assigning an energy penalty for “kinks” in the spacer configuration.
This spin model reproduces a few important properties of gemini surfactants as
observed in experiments, primarily the nonmonotonous dependence of the cmc on spacer
length for hydrophobic spacers and the formation of
The bead-spring MD approach discussed in Section II.B was extended as well to treat
gemini surfactants [36,37]. The only essential modification is the connection of head
groups in pairs by spacer chains. Like the tail chains, the spacers are made of oillike
particles connected to one another by harmonic springs, where only hydrophobic spacers
were studied. These MD simulations are able to reproduce the micellar shapes formed by
the m-s-m gemini surfactants—branched wormlike micelles and ring micelles—compared
to the spherical morphology formed by the corresponding monomeric surfactants (see
Fig. 7). A similar coarse-grained MD approach, along with a self-consistent-field
calculation, were applied to the more complex glucitol amine gemini surfactants, which
have flexible sugar side chains attached to the charged head groups [54]. The main
finding is a transition from cylindrical micelles to bilayers upon increasing pH, in accord
with experimental indications [55].
C.
Phase Behavior
The spacer length has an unusual effect also on the phase behavior of systems containing
m-s-m gemini surfactants. This fact has been mentioned already in Section I. For binary
surfactant-water mixtures, the regions in the concentration-temperature phase diagrams,
where single-phase hexagonal and lamellar phases are the stable state, become smaller as
s increases, vanish for s=10–12, and then are finite again for s=16 [16]. In ternary water-
oilsurfactant phase diagrams of the same surfactants, the size of the micro-emulsion
(single-phase) region increases with s and then decreases, with a maximum at s 10 [17].
These observations can be rationalized in the light of the packing considerations
discussed in Sections III.A and III.B [1]. As s increases from low values, the optimal area
per molecule a(s) increases and the packing parameter P of Eq. (6) decreases. Hence,
surfactant packing into bilayers, which are the building blocks of the lamellar and
hexagonal mesophases, becomes less and less favorable. As we have seen in the previous
sections, this effect is maximal for s =10–12, whereupon bilayers apparently can no
longer be stabilized. When s increases further, a decreases, and bilayers can form again.
This qualitative description agrees also with the experimental results for m-EOx-m gemini
surfactants. The observed monotonous increase in the phase-diagram region belonging to
the isotropic micellar phase [56] is in
Models of gemini surfactants 59
IV.
CONCLUSIONS AND OPEN QUESTIONS
ordering due to elongated head groups was theoretically addressed in the case of bilayer
membranes [58]. A recent work combining dichroism spectroscopy and atomistic MD
simulations has revealed orientational ordering of gemini surfactants in cylindrical
micelles [42]. More theoretical work is required to elucidate this issue. (3) The above two
aspects apply as well to double-tail surfactants (e.g., phospholipids) with large, elongated
head groups. What truly distinguishes gemini surfactants from double-tailed surfactants is
the fact that the spacer makes a soft link between the head groups. Containing at least
several chemical bonds, it allows a degree of conformational flexibility to the entire
molecule (e.g., with respect to the relative orientations of the two tails). This feature is
probably what allows gemini surfactants to form such uncommon structures as branched
micelles and ring micelles. In the MD simulations of Ref. 36, for example, the gemini
surfactants residing in a branching junction of a wormlike micelle were found to have
their two tails oriented in different directions. This property might also make gemini
surfactants serve as cross-linkers of regular micelles [59]. Hence, it seems that our
understanding of gemini surfactant self-assembly will be incomplete until we have a good
account of the interplay among various lateral organizations of these molecules at
surfaces.
Another important direction where there is substantial experimental information but
almost no theory is the dynamics and rheology of gemini surfactant solutions. This aspect
is particularly relevant to applications, as these solutions exhibit unusual and useful
rheological properties such as shear thickening at low volume fractions [1]. Moreover,
recently these properties have made micellar solutions of gemini surfactants a model
system for studying nonequilibrium behaviors such as shear thickening and ultraslow
relaxation [21]. We note that the dynamic issues and the issue of molecular organization
mentioned earlier may be closely related. It has been argued recently that the distinct
rheological behavior of wormlike micellar solutions (e.g., shear thickening) stems from
the formation and interlinking of ring micelles [60].
We hope that this review and the posed open questions will motivate further
theoretical studies of this class of fascinating and very useful self-assembling molecules.
ACKNOWLEDGMENTS
We are grateful to Raoul Zana for introducing us to the field of gemini surfactants and for
numerous discussions and suggestions. We also benefited from discussions with Igal
Szleifer. H.D. acknowledges support from the Israeli Council of Higher Education (Alon
Fellowship), and D.A. from the Israel Science Foundation under grant No. 210/02 and the
Alexander von Humboldt Foundation.
REFERENCES
5. Zhu, Y.P.; Masuyama, A.; Kobata, Y.; Nakatsuji, Y.; Okahara, M.; Rosen, M.J. J. Colloid Interf.
Sci. 1993, 158, 40–45.
6. Zana, R.; Berraou, M.; Rueff, R. Langmuir 1991, 7, 1072–1075.
7. Devinsky, F.; Lacko, I.; Imam, T. Acta Fac. Pharm. 1990, 44, 103.
8. De, S.; Aswal, VK.; Goyal, PS.; Bhattacharya, S. J. Phys. Chem. 1996, 100, 11,664–11,671.
9. Zana, R. J. Colloid Interf. Sci. 2002, 252, 259–261.
10. Grosmaire, L.; Chorro, M.; Chorro, C; Partyka, S.; Zana, R. J. Colloid Interf. Sci. 2002, 246,
175–181.
11. Israelachvili, J.; Mitchell, J.; Ninham, B.W. J. Chem. Soc. Faraday Trans. 2 1976, 72, 1525–
1568.
12. Israelachvili, JN. Intermolecular and Surface Forces, 2nd Ed.; Academic Press: San Diego,
CA, 1991.
13. Zana, R.; Talmon, Y. Nature 1993, 362, 228–230.
14. Danino, D.; Talmon, Y.; Zana, R. Langmuir 1995, 11, 1448–1456.
15. Bernheim-Groswasser, A.; Zana, R.; Talmon, Y. J. Phys. Chem. B 2000, 104, 4005–4009.
16. Alami, E.; Levy, H.; Zana, R.; Skoulios, A. Langmuir 1993, 9, 940–944.
17. Dreja, M.; Gramberg, S.; Tieke, B. Chem. Commun. 1998, 13, 1371–1372.
18. Kern, F.; Lequeux, F.; Zana, R.; Candau, SJ. Langmuir 1994, 10, 1714–1723.
19. In, M.; Bec, V.; Aguerre-Chariol, O.; Zana, R. Langmuir 2000, 16, 141–148.
20. Schmitt, V.; Schoseler, F.; Lequeux, F. Europhys. Lett. 1995, 30, 31–36.
21. Oelschlaeger, C.L.; Waton, G.; Candau, S.J.; Cates, M.E. Langmuir 2002, 18, 7265–7271.
22. Adamson, A.W.; Gast, A.P. Physical Chemistry of Surfaces, 6th Ed.; Wiley: New York, 1997.
23. Nagarajan, R.; Ruckenstein, E. Langmuir 1991, 7, 2934–2969.
24. Tanford, C. The Hydrophobic Effect; Wiley: New York, 1980.
25. Evans, D.F.; Ninham, B.W. J. Phys. Chem. 1983, 87, 5025–5032.
26. Widom, B. J. Chem. Phys. 1986, 84, 6943–6954.
27. Larson, R.G.; Scriven, L.E.; Davis, H.T. J. Chem. Phys. 1985, 83, 2411–2420.
28. Larson, R.G. J. Chem. Phys. 1992, 96, 7904–7918.
29. Stauffer, D.; Jan, N.; Pandey, R.B. Physica A 1993, 198, 401–409.
30. Stauffer, D.; Jan, N.; He, Y.; Pandey, R.B.; Marangoni, D.G.; Smith-Palmer, T. J. Chem. Phys.
1994, 100, 6934–6942.
31. Jan, N.; Stauffer, D. J. Phys. I (France) 1994, 4, 345–350.
32. Smit, B.; Hilbers, P.A.J.; Esselink, K.; Rupert, L.A.M.; van Os, N.M.; Schlijper, A.G. Nature
1990, 348, 624–625.
33. Smit, B.; Esselink, K.; Hilbers, P.A.J.; Vanos, N.M.; Rupert, L.A.M.; Szleifer, I. Langmuir
1993, 9, 9–11.
34. Palmer, B.J.; Liu, J. Langmuir 1996, 12, 746–753.
35. Fodi, B.; Hentschke, R. Langmuir 2000, 16, 1626–1633.
36. Karaborni, S.; Esselink, K.; Hilbers, P.A.J.; Smit, B.; Karthauser, J.; van Os, N.M.; Zana, R.
Science 1994, 266, 254–256.
37. Maiti, P.K.; Lansac, Y.; Glaser, M.A.; Clark, N.A. Langmuir 2002, 18, 1908–1918.
38. Watanabe, K.; Ferrario, M.; Klein, M.L. J. Phys. Chem. 1988, 92, 819–821.
39. Shelley, J.C.; Sprik, M.; Klein, M.L. Langmuir 1993, 9, 916–926.
40. Mackerell, A.D. J. Phys. Chem. 1995, 99, 1846–1855.
41. Maillet, J.B.; Lachet, V.; Coveney, P.V. Phys. Chem. Chem. Phys. 1999, 1, 5277–5290.
42. Oda, R.; Laguerre, M.; Huc, I.; Desbat, B. Langmuir 2002, 18, 9659–9667.
43. Gompper, G.; Schick, M. In Phase Transitions Critical Phenomena; Domb, C., Lebowitz, J.L.,
Eds; Academic Press: London, 1994.
44. In Micelles, Membranes, Micromulsions, Monolayers; Gelbart, W.M.; Ben-Shaul, A.; Roux,
D.; Eds.; Springer-Verlag: New York, 1994.
45. Safran, S.A. Statistical Thermodynamics of Surfaces, Interfaces, Membranes’, Addison-
Wesley: New York, 1994.
Models of gemini surfactants 63
I.
INTRODUCTION
Although this chapter covers several subjects quantitatively and in some detail, it is
not meant to be a comprehensive review of the subject. Thus, several articles are not
cited, for lack of space to discuss them adequately. The emphasis is on more recent
articles or articles which provide quantitative or unusual information. The goals are to
provide a general introduction to the subject and a brief review of research methods and
results. Gemini surfactants hold promise for use in many applications for improved
surface activity, efficiency, and performance (see Chapters 12 and 13). The surface
behavior of mixtures of conventional and gemini surfactants is not dealt with in this
chapter. Results on the synergism in monolayer formation and surface tension reduction
efficiency and effectiveness by such mixtures are found in Chapter 10.
II.
METHODS FOR DETERMINING SURFACE TENSION AND
ADSORPTION PROPERTIES
A.
Methods for Equilibrium Surface Tension
There are three major classes of methods for measuring the surface tension of air-liquid
interfaces and the interfacial tension of oil-water (liquid-liquid) interfaces: force methods,
shape methods, and pressure methods [5,6]. At the surface, there is a balance of the key
static forces: gravity forces (and centripetal forces, if applicable), static pressure forces,
and surface tension forces [7].
Two force methods, the Wilhelmy plate method and the Du Noüy method, are used
most often to measure surface or interfacial tensions and often the contact angles [5,8]. It
is best to use them by having the force measured automatically and continuously without
perturbing the surface, as would happen when the plate is pulled by a wire. In the Du
Noüy method, one uses a platinum ring instead of a plate and uses an empirical correction
factor, accounting or the rise of the liquid at the interior space of the ring [4], which may
lead to inaccurate results, especially for oil-water interfaces.
To avoid artifacts from surface active impurities, the surfactant must be purified to the
furthest extent possible, via recrystallization or other procedures. Additionally, the
monolayer should be aspirated (using a low-vacuum liquid suction) to ensure that the
equilibrium surface tension returns to the same value, which usually indicates surface
chemical purity [9,10].
The shape methods, such as the pendant-drop method, the emerging-bubble method,
the sessile-drop method, and the captive-bubble method, rely on gravity to deform the
drop or bubble [11]. The shape is fit to a model using the axisymmetric drop-shape
analysis, which relies on the general Laplace-Young equation [5,7]. These methods are
generally more accurate than the force methods because they do not involve a solid-
liquid-gas contact line and a contact angle θ, which is uncertain and a source of error, but
no surface aspiration can be used for surface purification. The spinning-drop or spinning-
bubble method is based on a balance of centrifugal and surface tension forces [12,13]. In
pressure methods, as in the bubble surfactometer, a small spherical air-water surface,
open to the atmosphere, is used. The pressure difference is measured across the curved
Gemini surfactants 66
interface, and the surface tension γ is found from the Laplace-Young equation for
spherical surfaces [14,15].
B.
Methods for Determining Surface Densities and Probing Adsorbed
Layers
The surface excess density Γ (mol/m) is close to the surface density of adsorbed
molecules of surface-active solutes. The quantities Γi of surfactants i=1,2,…are related to
the variation of the equilibrium surface tension with the chemical potentials µi via the
Gibbs adsorption equation, or “isotherm” [5,8,16]:
(1)
For one nonionic or zwitterionic surfactant and an ideal bulk solution (or nonideal, but
with the activity coefficient f [17] constant or varying slowly with concentration),
(2)
(3)
where Γm is the maximum surface density or adsorption capacity and KL is the adsorption
equilibrium constant. Equations (2) and (3) yield the Gibbs-Szyszkowski equation, which
was first found empirically [5]:
(4)
From Eqs. (3) and (4), one gets the surface equation of state
(5)
The maximum value, Γmax, of Γ found from Eq. (2) is often considered to be equal to Γm.
Usually, however, Γmax < Γm. For most molecules with one straight hydrocarbon chain,
Γm≤8×10−6 mol/m2 or a≥0.2 nm2/molecule (the area per oriented crystalline, or all trans
hydrocarbon chains), where a= (NAΓm)−1 (NA is Avogadro’s number). For molecules with
two hydrocarbon chains, such as the gemini surfactants, Γm≤4×10−6 mol/m2 or a≥0.4 nm2/
molecule. The minimum area can be much higher than 0.4 nm2 for gemini surfactants
with long spacers, nonstraight chains, or bulky head groups.
Adsorption and surface tension behavior of gemini surfactants 67
For ionic surfactant salts, the equations depend on whether the surfactant is a strong or
weak electrolyte, on the surfactant concentration C, and the salt concentration Cs of any
surface-inactive electrolytes present. The general equation is [see Eq. (2)]
(6)
where n is a function of C, CS, and the number of dissociating ionic species per molecule
[7,18]. Equation (4) is then modified to yield
(7)
With no added salt, for a 1:1 strong electrolyte, R+X−, n=2; for a 2:1 strong electrolyte,
X−R+≡R+X−, n=3; and so on. For high concentrations of salt, Cs>>C (“swamping”
concentrations of supporting electrolyte), n=1 in all cases. For intermediate
concentrations, 1<n<2 (for conventional surfactants) or 1<n<3 (for gemini surfactants)
[19]. As with nonionic surfactants, Γ1 can be determined directly from γ versus In C
curves or from fitting to a model. Several models have been reviewed recently [18,20].
Various methods can be used to directly probe the adsorbed surface monolayers. The
first method used involved radiotracers [21–23]. This method has not been used for
gemini surfactants. Optical ellipsometry can be used to measure surface densities from
the intensity and state of polarization of light reflected from the aqueous surface [24,25].
This method is not chemically specific, requires an independent measurement of the
refractive index increment, dn/dc, and is often not sensitive enough for determining the
thickness and refractive index of the adsorbed monolayer unambiguously. One may also
use Brewster angle microscopy as an imaging technique for obtaining quantitative or
qualitative information [26]. A related method is optical reflectometry, in which the
reflected intensity is measured [27].
Infrared reflection absorption spectroscopy (IRRAS) is a chemically specific method
which can be used to probe not only the surface density but also the chain conformation
of adsorbed molecules [25,28–31]. The method works for insoluble monolayers and for
soluble monolayers, usually better if the bulk concentration does not exceed 10 mM [30].
A neutron reflection technique has been used recently, primarily by the groups of Thomas
and Penfold [32,33] for determining adsorbed monolayer thickness and surface densities
and also for probing counterions [34]. Finally, the surface potential of aqueous ionic
surfactant solutions can be probed by electrodes [35].
C.
Methods for Measuring Dynamic Surface Tensions or Surface
Densities
Many phenomena involving surfactants, such as foaming, emulsion or drop formation,
and coating flows are affected by the time-dependent surface tension or “dynamic surface
tension” (DST). Whereas the surface tensions of pure liquids or solutions of surface
inactive molecules are believed or expected to be established, or to equilibrate, extremely
fast, within 10−8−10−5 s, the surface tensions of aqueous surfactants equilibrate at
timescales ranging from 10−5 to 105 s. The equilibration times depend on the surfactant
Gemini surfactants 68
III.
EQUILIBRIUM RESULTS AT AIR-WATER INTERFACES
A.
Nonionic and Zwitterionic Surfactants
Fewer studies are available for nonionic and zwitterionic than for cationic gemini
surfactants [37–41]. Interfacial properties were used to infer the effect of the nature and
position of the spacer and of the alkyl chain length on the properties of sugar-based
nonionic gemini surfactants (Table 1). Surface tensiometry techniques have been used to
study the pH-induced micelle-to-vesicle transition observed for hexane-1,6-
bis(hexadecyl-1′-deoxyglycityl-amine), a nonionic sugar-amine gemini surfactant [38].
Having no net charge, zwitterionic heterogemini surfactants adsorb similarly to nonionic
surfactants [39,40]. The tension-based critical micelle concentrations (cmc’s) were found
to be consistent with those from fluorescence, of the order of 10−5 M. The inferred values
of the surface area per molecule, between 0.2 and 0.3 nm , were lower, however, than
would be expected for close-packed monolayers, probably due to strong electrostatic
interactions between the zwitterionic head groups [40] or to surface aggregation [39]. The
addition of NaBr was found to alter the surface activity considerably [40].
The interfacial behavior of two lactose derivatives having, in sequence, a long
hydrocarbon chain, a nonionic polar head, a long hydrocarbon spacer, and as a second
Adsorption and surface tension behavior of gemini surfactants 69
B.
Ionic Surfactants
For heterogemini surfactants in which one head group was a nonionic methyl-capped
polyoxyethylene chain (of molecular weight 350, 550, or 750 g/mol) and the other was an
anionic sodium sulfate group, the increase in the surface tension at the cmc with the
polyoxyethylene chain length was attributed to steric “crowding” of nonionic head
groups at the air-water interface [42]. For two anionic gemini surfactants, sodium N-
dodecanoyl-N-methyltaurinate (SDMT) and sodium decyloxyacetate (SDOA), evidence
was found of steric hindrance to micellization and surface packing, attributed
TABLE 1 Summary of Equilibrium Surface
Tension Data for Nonionic, Zwitterionic, and
Heterogemini Surfactants
Surfactant Surfactant Surface T, ºC Reported Cmc Dimer vs. Variable Refer
type tension interfacial M monomer? spacer ence
measurement properties length?
method
Nonionic alkyl glucoside maximum 25 γ(C), cmcγ, 10−2 Yes Yes 37
derivatives bubble Γm, a,
pressure ∆Gomic,
∆Goads,
PC20,
cmc/C
HLB
Nonionic sugar-amine Lauda drop 45 γ(pH, C=1 10−3 No No 39,40
derivative tensiometer mg/ml),
γ(pH=3.0,
C,) cmcγ, a
Zwitterionic C3-PO4-(CH2)2- Du Noüy ring 20 γ(C,Cs), (10−5) Yes No 52
N+ (cH2)2C3 γ
cmc ,Γm, a,
Dissymmetric Bis(quat) Wilhelmy 30–40 γ(C, T), (10−4) Yes No 42
plate Tk, cmcγ,
Γm,a,
∆Gºmic,
∆Gº ads, β
Asymmetric Nonionic/anionic Du Noüy ring 20 (wt%), 10−3 42 Yes No
cmcγ, Γm,
a)
Asymmetric Lactose stirrup 25,37,50 γ(C), cmcγ, (10−3) No Yes 41
derivatives detachment Γ m, a
Gemini surfactants 70
technique
γ
cmc : tension based cmc.
∆Gºmic: Standard free energy of micellization.
∆Gºads: Standard free energy of adsorption.
PC20: negative logarithm of the surfactant concentration required to decrease the surface tension by 20mN/m.
cmc/C: relative effect of micellization to adsorption.
HLB: hydrophilic lipophilic balance.
Tk: Krafft temperature.
β: degree of countering binding.
to hydrogen-bonding between the amide and protonated carboxylate polar head groups
[43]. Most gemini surfactants studied to date are cationic: 16–4– 16, a dimer of CTAB
(Cetyl Trimethylammonium Bromide) [44]; c12-s-c12 (where c12 is cyclododecyl) with
s=3, 4, and 6 [45]; a series of 24 dicationics having a 2-butynyl or a butanediyl spacer
and an alkyl chain from octyl to octadecyl [46]; bis(quaternary ammonium halide) with
heteroatomic spacers [47,48]; and 12-s-12 [49–51]. Of particular interest is a neutron
reflection study [32] of adsorbed monolayers of a series of 12-s-12 surfactants, with s=3,
4, 6, 12, and of 12-xylyl-12. Measurements of the surface areas per molecule at the cmc
were compared to the indirect results from the Gibbs equation [Eq. (6)]. For the gemini
surfactants with flexible methylene spacers, a value of n=2 was found to agree better
with the neutron reflectivity data than the value of n=3, suggesting that the dicationic
gemini surfactants may act as chelating agents. For gemini surfactants with a rigid xylyl
spacer, a value of n=3 was found, as expected for a fully dissociated gemini surfactant.
Further neutron reflectivity studies suggest that beneath the adsorbed surfactant
monolayer lies a loose sublayer of surfactant molecules, oriented as if in a bilayer
structure [33].
The surface area per molecule was determined for 12-s-12 cationic geminis with
hydrophobic spacer chains to range from 1 to 2.4 nm /molecule and show a maximum
with the spacer carbon number 5 (Fig. 1) [51]. The spacer is presumed to be located at the
interface for s<10 and at the air side for s≥10. These results correlate with the formation
in solution of elongated micelles at small s, spherical micelles at intermediate, s, and
vesicles at large s. 12-EOx-12 geminis with hydrophilic CH2CH2(OCH2CH2)x spacer
chains have values of a which increase with the spacer chain length (see Fig. 1).
The interfacial properties of the heterogemini surfactant 12–2–14 were similar to those
found for the symmetric gemini counterparts 12–2–12 and 14–2–14 [52].
Several cationic gemini surfactants Nα,Nω-bis(Nα-lauroyl-arginine)α,ω
alkylidenediamide [Cs(LA)2] were studied [53–56] (see Chapter 2, Scheme 7).
Comparison of the equilibrium surface tensions (Fig. 2) of Cs(LA)2 to those of the
monomeric surfactant LAM (Nα-lauroyl-arginine methyl ester) shows that the gemini
surfactants have a surface tension-based cmc which is much smaller than that of LAM by
two or three orders of magnitude [57,58]. Use of other methods, primarily ion activity (of
Cl−), conductimetry, and fluorescence, show that the aggregates beyond the tension cmc,
or first cmc (cmc1), are not traditional micelles [58]. These methods revealed a higher
second cmc (cmc2), beyond which traditional micelles capable of solubilizing
fluorophores and having substantial counterion binding were present. The minimal
Adsorption and surface tension behavior of gemini surfactants 71
observed surface tensions for LAM and gemini surfactants were the same, indicating
similar surface densities of hydrocarbon chains and polar
m3/mol, or four orders of magnitude higher. For mixed LAM/C6(LA)2 solutions, one can
estimate their equilibrium surface tensions and surface densities of the individual
components using the ideal adsorbed solution theory [59,60]. Even though Γm1>Γm2,
because KL2>KL1, one would expect that C6(LA)2 would be the dominant component at
the surface when mixtures are used.
For ionic surfactants, the pseudononionic model is generally inadequate. Using ionic
surfactant models, one can describe not only γ and Γ but also the
surface potential Φ0. For this model, one uses the Langmuir or Frumkin isotherm as a
basis. One may use a counterion binding fraction β by analogy to micelles. Using β=0.7
and the equations [19],
(8)
(9)
one can develop a model with the equilibrium constant K and the energy of interaction A
as adjustable parameters. Here, ΓGC is the Gouy-Chapman surface density (for a 2:1
surfactant in the presence of a 1:1 salt with a common ion) corresponding to the net
surface charge, z is the adsorbed surfactant ion valence, ε is the permittivity of free space,
εr is the dimensionless dielectric constant of the fluid, and F is Faraday’s constant. This
model fits the surface tension data for LAM and C6(LA)2 fairly well, and the calculated
data on Γ and Φ0 fairly; see Figure 3 [19]. Using ellipsometry and IRRAS, it was found
that the surface densities (in mg/m2) were the same for LAM and C6(LA)2 above their
respective cmc’s (cmc1 for the gemini surfactant) (Fig. 4) [25]. Both techniques showed
that C6(LA)2 produces its largest surface density at
C.
Spread Monolayer Studies
For a series of (nonionic) double-chain diols with chain length m=10–18 and spacer
length s=2–6, more tightly packed monolayers were inferred as m increased and as s
decreased [62]. The presence of an unsaturated (double) bond in the middle of the tails
decreases the packing efficiency in the monolayer. The above results were obtained with
the standard Π-a method in a Langmuir trough from 10°C to 50°C. Similar objectives
were sought with diols or triols and trimeric surfactants with sulfonate groups [63]. A
series of nonionic geminis with disaccharide trehalose spacers are insoluble for chain
lengths of 12 carbons or larger [64]. The surface pressure of spread monolayers starts
increasing substantially for a<1.2 nm2/molecule. The collapse pressure is about 50 mN/m
at a=0.5 nm2/molecule and increases slightly
IV.
DATA AND MODELS FOR DYNAMIC SURFACE TENSION
Dynamic surface tensions for gemini surfactant solution have been obtained with the
emerging bubble (often called pendant bubble), the bubble method at
(10)
For certain ionic heterogemini surfactants, the dynamic surface tensions equilibrated
within 10–100 s [71]. Alami et al. [42,71] used an empirical equation by Hua and Rosen
[72] to correlate the data. Then they used Ward and Tordai’s integral expression to relate
Γ(t) to D, C, and t [70]. Various approximate equations were used to infer the possible
presence of an adsoption barrier, which makes equilibration slower than predicted by the
Adsorption and surface tension behavior of gemini surfactants 79
diffusion-controlled model. The above models do not account explicitly for electrostatic
effects, which are important for ionic surfactants.
The dynamic surface tensions of C6(LA)2 solutions were determined with the bubble
method at constant area (Fig. 8) or pulsating area [73]. The gemini surfactants
equilibrated faster than monomeric LAM at the same molar concentration, because of
their higher surface activity, as expressed by the value of the adsorption equilibrium
constant KL. The results showed that micelles or aggregates increase the net adsorption
rate and thus contribute to the dynamic surface tension but not to the equilibrium surface
tension. This indicates that micelles or aggregates can dissolve fast and replenish the
depleted reservoir of adsorbable monomeric surfactants [73]. Similar results were
obtained with solutions of disulfur betaine-derivative gemini surfactants [74]. Using
pseudononionic adsorption models, it was inferred that there was a strong barrier to
adsorption, which was slower than the diffusion-controlled prediction.
V.
APPLICATIONS TO FOAMS
Foam stability and breakup depend on a series of complex phenomena, starting from
hydrodynamic liquid drainage, aqueous film thinning, and bubble coalescence. These
phenomena are affected by equilibrium surface tension, the equilibrium surface equation
of state, Gibbs elasticity, surface charge, film disjoining pressure, and also complex
surface deformation and dynamic surface tension properties [75,76]. The addition of
appropriate surfactants allows the film lifetime to be increased from a few seconds to
weeks. Thus, much effort has been spent studying the influence of the surfactant
molecular structure on foam formation and stability [77]. Foaming is also, in principle,
related to the dynamic adsorption and surface tension of aqueous surfactants. Indeed,
when the solutions are shaken, a large surface area free of surfactant is created. Then, the
molecules adsorb from the bulk solution, and the foam film surface is covered by
monolayers which protect against rupture via an electrostatic, steric, or other repulsion
mechanism.
The ability of LAM and Cs(LA)2 to form stable foams was measured from the
maximum foam heights with solutions confined in glass vials and shaken vigorously. The
data correlated best when plotted as dimensionless foam height, H/H0, where H0 is the
maximum possible height, versus the dimensionless concentration C/cmc1 for LAM or
C/cmc2 for Cs(LA)2. Gemini surfactants were more efficient foam stabilizers than LAM.
The foam decay times for the foam height to drop by 50% correlated strongly with the
reduced concentration (Fig. 9), indicating that the geminis are 20 times more efficient
than LAM. The foam decay times correlate well with the tension equilibration times or
the time for the surface tension to drop by 50% of its equilibrium decrease (Fig. 10). The
two timescales were fit empirically to a straight line, log , where for
LAM, A=4.16 and B=−0.03, and for C6(LA)2, A=4.07 and B=−0.002, with t in seconds.
Although the
Gemini surfactants 80
for 0.1 wt% surfactant, above the cmc’s [78–82]. Foam stability reached a maximum at
the cmc. The initial foam height increased as the surface tension at the cmc decreased,
indicating a strong correlation between foaming ability and stability with molecular
structure, in particular the presence of one, two, or three chains, and the spacer type and
size. Foam stability was poor when the solubility of the gemini surfactants was low.
For surfactants with two ammonium groups with different spacer chains and
hydrophilic [83,84] and butenylene or butynylene groups [85] at 0.1 wt%, foam stability
is strongly affected by the nature of the spacer group. The gemini surfactants showed
much better foaming ability and foam stability than the conventional ones, and cationic
geminis were better than anionic geminis.
For thin liquid films stabilized by the cationic gemini surfactant 12–2–12, in the dilute
and in the semidilute regimes, the disjoining pressure was found to depend on the film
thickness, to be positive or negative, and to affect foam stability [86]. The gemini
surfactant could stabilize thin films efficiently at low concentration, in contrast with the
corresponding monomeric surfactant. The formation of stable Newton black films was
observed in the presence of an added electrolyte. In the semidilute concentration regime,
oscillatory structural forces were observed with varying film thickness when wormlike
micelles were present in the solution, as found for polyelectrolytes but not neutral
polymers [87–89]. These forces were not observed in the presence of added salt. This
suggests that in the absence of salt, there are strong electrostatic forces between the
charged aggregates, which are “screened” considerably by added salt.
VI.
RESULTS AT OIL-WATER INTERFACES AND EMULSIONS
solution. The behavior at oil-water interfaces was often related to that at air-water
interfaces.
VII.
RESULTS FOR SOLID-WATER INTERFACES
VIII.
CONCLUSIONS
This chapter covers certain experimental methods and theoretical foundations for
studying and interpreting the interfacial adsorption behavior of gemini surfactants. Key
results on adsorption at air-water, oil-water, and solid-water interfaces are reviewed. The
results are examined at several levels, from the empirical and qualitative to the rigorous
and quantitative. The behavior of gemini surfactants has received much attention over the
last 10 years or so, because of their relative novelty and promise of higher effectiveness
and efficiency than those of monomeric surfactants. This bodes well for applications in
foaming, wetting, emulsions, and biological/biotechnological applications. Various novel
molecular architectures have been synthesized, differing with regard to the polar head
groups (same or different), the hydrocarbon chains (same or different size), and the types
and sizes of the spacer. Future experimental and theoretical work should help establish
further our understanding of the effects of these molecular properties on the phase,
aggregation, and interfacial behavior.
ACKNOWLEDGMENT
This research was supported in part (E.I.F. and A.J.P.) by the National Science
Foundation (grant No. CTS 01–35317).
REFERENCES
17. Smith, J.M.; VanNess, H.C.; Abbott, M.M. Introduction to Chemical Engineering
Thermodynamics, 6th Ed.; McGraw-Hill: New York, 2001.
18. Prosser, A.J.; Franses, E.I. Colloids Surfaces A 2001, 171, 1.
19. Prosser, A.J.; Franses, E.I. J. Colloid Interf. Sci. 2001, 240, 590.
20. Fainerman, V.B.; Lucassen-Reynders, E.H.; Miller, R. Colloids Surfaces A 1998, 143, 141.
21. Tajima, K.; Muramatsu, M.; Sasaki, T. Bull. Chem. Soc. Jpn. 1970, 43, 1991.
22. Tajima, K. Bull. Chem. Soc. Jpn. 1970, 43, 3063.
23. Tajima, K. Bull. Chem. Soc. Jpn. 1971, 44, 1767.
24. DeFeijter, J.A.; Benjamins, J.; Veer, F.A. Biopolymers 1978, 17, 1759.
25. Walsh, C.B.; Wen, X.; Franses, E.I. J. Colloid Interf. Sci. 2001, 233, 295.
26. Henon, S.; Meunier, J.J. Chem. Phys. 1993, 95, 9148.
27. Tompkins, H.G.; McGahan, W.A. Spectroscopic Ellipsometry and Reflectometry, Wiley: New
York, 1999.
28. Dluhy, R.A.; Cornell, D.G. J. Phys. Chem. 1985, 89, 3195.
29. Mendelsohn, R.; Brauner, J.W.; Gericke, A. Annu. Rev. Phys. Chem. 1995, 46, 305.
30. Prosser, A.J.; Franses, E.I. Langmuir 2002, 18, 9234.
31. Born, M.; Wolf, E. Principles of Optics; 4th Ed, Pergamon Press: New York, 1970.
32. Li, Z.X.; Dong, C.C.; Thomas, R.K. Langmuir 1999, 15, 4392.
33. Li, Z.X.; Dong, C.C.; Wang, J.B.; Thomas, R.K.; Penfold, J. Langmuir 2002, 18, 6614.
34. Su, T.J.; Thomas, R.K.; Penfold, J. Langmuir 1997, 73, 2133.
35. MacRitchie, F. Chemistry at Interfaces; Academic Press: San Diego, CA, 1990.
36. Pinazo, A.; Wen, X.; Liao, Y.-C.; Prosser, A.J.; Franses, E.I. Langmuir 2002, 18, 8888.
37. Castro, M.J.L.; Kovensky, J.; Cirelli, A.F. Langmuir 2002, 18, 2477.
38. Bergsma, M.; Fielden, M.L.; Engberts, J.B.F.N. J. Colloid Interf. Sci. 2001, 243, 491.
39. Seredyuk, V.; Alami, E.; Nyden, M.; Holmberg, K.; Peresypkin, A.V.; Menger, F.M. Langmuir
2001, 17, 5160.
40. Seredyuk, V.; Alami, E.; Nydén, M.; Holmberg, K.; Peresypkin, A.V.; Menger, F.M. Colloids
Surfaces A 2002, 203, 245.
41. Gouzy, M.-F.; Guidetti, B.; André-Barres, C.; Rico-Lattes, I.; Lattes, A.; Vidal, C.J. Colloid
Interf. Sci. 2001, 239, 517.
42. Alami, E.; Holmberg, K.; Eastoe, J. J. Colloid Interf. Sci. 2002, 247, 447.
43. Tsubone, K.; Rosen, M.J. J. Colloid Interf. Sci. 2001, 244, 394.
44. Oliviero, C; Coppola, L.; La Mesa, C; Ranieri, G.A.; Terenzi, M. Colloids Surfaces A 2002,
201, 247.
45. Jenkins, K.M.; Wettig, S.D.; Verrall, R.E. J. Colloid Interf. Sci. 2002, 247, 456.
46. Menger, F.M.; Keiper, J.S.; Azov, V. Langmuir 2000, 16, 2062.
47. Rosen, M.J.; Mathias, J.H.; Davenport, L. Langmuir 1999, 15, 7340.
48. Song, L.; Rosen, M.J. Langmuir 1996, 12, 1149.
49. Esumi, K.; Miyazaki, M.; Arai, T.; Koide, Y. Colloids Surfaces A 1998, 735, 117.
50. Dam, Th.; Engberts, J.B.F.N.; Karthäuser, J.; Karaborni, S.; van Os, N.M. Colloids Surfaces A
1996, 115, 41.
51. Alami, E.; Beinert, G.; Marie, P.; Zana, R. Langmuir 1993, 9, 1465.
52. Sikiric, M.; Primožič, I.; Filipović-Vinceković, N.J. Colloid Interf. Sci. 2002, 250, 221.
53. Infante, R.; Garcia Dominguez, J.; Erra, P.; Julia, R.; Prats, M. Int. J. Cosmet. Sci. 1984, 6, 275.
54. Solans, C.; Infante, R.; Azemar, T.; Wärnheim, T. Prog. Colloid Polym. Sci. 1989, 79, 70.
55. Solans, C.; Pes, M.A.; Azemar, T.; Infante, M.R. Prog. Colloid Polym. Sci. 1990, 81, 144.
56. Pérez, L.; Torres, V.L.; Manresa, A.; Solans, C.; Infante, M.R. Langmuir 1996, 12, 5296.
57. Pérez, L.; Pinazo, A.; Rosen, M.J.; Infante, M.R. Langmuir 1998, 14, 2307.
58. Pinazo, A.; Wen, X.; Pérez, L.; Infante, M.R. Langmuir 1999, 15, 3134.
59. Franses, E.I.; Siddiqui, F.A.; Ahn, D.J.; Chang, C.-H.; Wang, N.-H.L. Langmuir 1995, 11,
3177.
60. Siddiqui, F.A.; Franses, E.I. Langmuir 1996, 12, 354.
Gemini surfactants 90
61. Prosser, A.J. Thermodynamics of equilibrium adsorption and surface tension of single and
binary ionic surfactant systems, PhD thesis, Purdue University, West Lafayette, IN, 2002.
62. Sumida, Y.; Masuyama, A.; Oki, T.; Kida, T.; Nakatsuji, Y.; Ikeda, I.; Nojima M. Langmuir
1996, 12, 3986.
63. Sumida, Y.; Oki, T.; Masuyama, A.; Maekawa, H.; Nishiura, M.; Kida, T.; Nakatsuji, Y.; Ikeda,
I.; Nojima, M. Langmuir 1998, 14, 7450.
64. Menger, F.M.; Mbadugha, B.N.A. J. Am. Chem. Soc. 2001, 123, 875.
65. Chen, X.; Wang, J.; Shen, N.; Luo, Y.; Li, L.; Liu, M.; Thomas, R.K. Langmuir 2002, 18, 6222.
66. Buijnsters, P.J.J.A.; Garcia Rodríguez, C.L.; Willighagan, E.L.; Sommerdijk, N.A.J.M.;
Kremer, A.; Camilleri, P.; Feiters, M.C.; Nolte, R.J.M.; Zwanenburg, B. Eur. J. Org. Chem.
2002, 2002, 1397.
67. Gao, T.; Rosen, M.J. J. Am. Oil Chem. Soc. 1994, 71, 771.
68. Rosen, M.J.; Song, Li. D.J. Colloid Interf. Sci. 1996, 179, 261.
69. Ferri, J.K.; Stebe, K.J. Colloids Surfaces A 1999, 156, 567.
70. Ward, A.F.H.; Tordai, L.J. Chem. Phys. 1946, 14, 453.
71. Alami, E.; Holmberg, K. J. Colloid Interf. Sci. 2001, 239, 230.
72. Hua, X.Y.; Rosen, M.J. J. Colloid Interf. Sci. 1988,124, 652.
73. Pinazo, A.; Perez, L.; Infante, M.R.; Franses, E.I. Colloids Surfaces A 2001, 189, 225.
74. Pinazo, A.; Diz, M.; Solans, C.; Pes, M.A.; Erra, P.; Infante, M.R. J. Am. Oil Chem. Soc. 1993,
70, 37.
75. Prud’homme, R.K.; Khan, S.A. In Foams: Theory, Measurements, and Applications;
Prud’homme, R.K., Khan, S.A., Eds.; Marcel Dekker: New York, 1996.
76. Langevin, D. In Foams and Emulsions; Sadoc, J.F., Rivier, N., Eds.; Kluwer: Boston, 1999.
77. Wilde, P.J. Curr. Opin, Colloid Interf. Sci. 2000, 5, 176.
78. Zhu, Y.; Masuyama, A.; Okahara, M.J. Am. Oil Chem. Soc. 1990, 67, 459.
79. Zhu, Y.; Masuyama, A.; Okahara, M.J. Am. Oil Chem. Soc. 1991, 68, 268.
80. Zhu, Y.; Masuyama, A.; Okahara, M.J. Am. Oil Chem. Soc. 1991, 68, 539.
81. Zhu, Y.; Masuyama, A.; Kirito, Y.; Okahara, M.; Rosen, M.J. J. Am. Oil Chem. Soc. 1992, 69,
626.
82. Yano, W.; Kimura, W.J. Jpn. Oil Chem. Soc. 1962, 11, 138.
83. Kim, T.-S.; Kida, T.; Nakatsuji, Y.; Hirao, T.; Ikeda, I. J. Am. Oil Chem. Soc. 1996, 73, 907.
84. Tatsumi, T.; Zhang, W.; Kida, T.; Nakatsuji, Y.; Ono, D.; Takeda, T.; Ikeda, I. J. Surfact.
Detergents 2000, 3, 167.
85. Tatsumi, T.; Zhang, W.; Nakatsuji, Y.; Miyake, K.; Matsushima, K.; Tanaka, M.; Furuta, T.;
Ikeda, I.J. Surfact. Detergents 2001, 4, 271.
86. Espert, A.; von Klitzing, R.; Poulin, P.; Colin, A.; Zana, R.; Langevin, D. Langmuir 1998,14,
4251.
87. Bergeron, V.; Langevin, D.; Asnacios, A. Langmuir 1996, 12, 1550.
88. Asnacios, A.; Espert, A.; Colin, A.; Langevin, D. Phys. Rev. Lett. 1997, 78, 4974.
89. Milling, A. J. Chem. Phys. 1996, 100, 8986.
90. Menger, F.M.; Littau, C.A. J. Am. Chem. Soc. 1993, 115, 10083.
91. Rosen, M.J.; Liu, L. J. Am. Oil Chem. Soc. 1996, 73, 885.
92. Zana, R.J. J. Colloid Interf. Sci. 2002, 246, 182.
93. Devinsky, F.; Lacko, I.; Imam, T. J. Colloid Interf. Sci. 1991, 143, 336.
94. Kunieda, H.; Masuda, N.; Tsubone, K. Langmuir 2000, 16, 6438.
95. Dreja, M.; Tieke, B. Langmuir 1998, 14, 800.
96. Dreja, M.; Pyckhout-Hintzen, W.; Mays, H.; Tieke, B. Langmuir 1999, 15, 391.
97. Wiaącek, A.E.; Chibowski, E.; Wilk, K. Colloids Surfaces B 2002, 25, 243.
98. Chorro, C.; Chorro, M.; Dolladille, O.; Partyka, S.; Zana, R. J. Colloid Interf. Sci. 1998,199,
169.
99. Chorro, M.; Chorro, C.; Dolladille, O.; Partyka, S.; Zana, R. J. Colloid Interf. Sci. 1999, 210,
134.
Adsorption and surface tension behavior of gemini surfactants 91
100. Grosmaire, L.; Chorro, C.; Chorro, M.; Partyka, S.; Zana, R. J. Colloid Interf. Sci. 2001, 243,
525.
101. Grosmaire, L.; Chorro, M.; Chorro, C; Partyka, S.; Boyer, B. Thermochim. Acta 2001, 379,
261.
102. Grosmaire, L.; Chorro, M.; Chorro, C.; Partyka, S. J. Colloid Interf. Sci. 2001, 242, 395.
103. Manne, S.; Schaffer, T.E.; Huo, Q.; Hansma, P.K.; Morse, D.E.; Stucky, G.D.; Aksay, I.A.
Langmuir 1997, 13, 6382.
104. Fielden, M.L.; Claesson, P.M.; Verrall, R.E. Langmuir 1999, 15, 3924.
105. Infante, M.R.; Diz, M.; Manresa, A.; Pinazo, A.; Erra, P. J. Appl. Bacteriol. 1996, 81, 212.
106. Esumi, K.; Goino, M.; Koide, Y. Colloids Surfaces A 1996, 118, 161.
107. Esumi, K.; Takeda, Y.; Goino, M.; Ishiduki, K.; Koide, Y. Langmuir 1997, 13, 2585.
108. Esumi, K.; Takeda, Y.; Koide, Y. Colloids Surfaces A 1998, 135, 59.
109. Esumi, K.; Uda, S.; Goino, M.; Ishiduki, K.; Tsuneo, S.; Hiroshi, F.; Koide, Y. Langmuir
1997, 13, 2803.
110. Esumi, K.; Toyoda, A.; Goino, M.; Suhara, T.; Fukui, H.; Koide, Y. J. Colloid Interf. Sci.
1998, 202, 377.
111. Li, F.; Rosen, M.J. J. Colloid Interf. Sci. 2000, 224, 265.
5
State of Gemini Surfactants in Solution at
Concentrations Below the cmc
RAOUL ZANA Institut C.Sadron (CNRS-ULP), Strasbourg, France
I.
INTRODUCTION
This chapter considers the behavior of gemini (gemini) surfactants in aqueous solution in
the submicellar range of concentration that is at concentrations below the critical
micellization concentration (cmc). The understanding of this behavior is extremely
important for the interpretation of different types of results. Consider, for instance, the
determination of the surface area a occupied by one surfactant at the air-water interface
from the variation of the surface tension γ with the surfactant concentration C. On the
basis of Gibbs law, the relationship between a and the surface excess Γ is:
(1)
In Eq. (1), R is the gas constant, T is the absolute temperature, NA is Avogadro’s number,
and n is a constant that takes the values 1 for nonionic surfactants, 2 for univalent-
univalent ionic surfactants, and 3 for divalent-univalent ionic surfactants. The ionic
gemini surfactants investigated are all of the divalent-monovalent type or conversely. The
value n=3 must therefore be used if at C<cmc, the gemini surfactant is completely
ionized in the solution from which adsorption occurs, whereas the value n=2 should be
used if the gemini surfactant ion binds one counterion (ion pairing). This uncertainty
concerning the value of n has been a real problem in the interpretation of surface tension
data for gemini surfactants (see Sec. II).
Another problem, as important as the previous one but still more complex, concerns
the formation of small premicellar aggregates at C<cmc. Premicellar aggregation
(premicellization) has been much studied for conventional surfactants. It appears to occur
more readily with gemini surfactants than with conventional ones, perhaps because of the
proximity of the two alkyl chains in the gemini surfactant molecule. Premicellar
aggregation can substantially perturb the determination of the value of a, resulting in
apparent values of this quantity. It also affects the value of the cmc and of the
thermodynamic quantities extracted from the cmc.
A last problem, which is specific to gemini surfactants at C<cmc, concerns the
possible existence of contacts between the two alkyl chains of a dimeric surfactant,
driven by the hydrophobic interaction between these chains. The occurrence of such
contacts will affect the thermodynamics of solutions of gemini surfactants by reducing
State of gemini surfactants in solution at concentrations below the cmc 93
the free energy of transfer of the dimeric surfactant from the aqueous phase to the micelle
and, thus, increasing the cmc.
The above three effects (ion pairing, premicellar aggregation and intra-molecular
interaction between alkyl chains) are successively discussed in this chapter. A last section
reviews some studies that relate to the conformation of gemini surfactants at C<cmc.
Because many of the studies that are reviewed in this chapter deal with the cationic
gemini surfactants alkanediyl-α,ω-bis(alkyldimethylammonium bromide), we refer to
these surfactants by the usual abbreviation m-s-m, where m and s are the carbon numbers
of the alkyl and alkanediyl groups (see Chapter 1).
II.
ION-PAIR FORMATION IN THE SUBMICELLAR RANGE OF
CONCENTRATION
The possibility of ion pairing in submicellar solutions of gemini surfactants arose in the
interpretation of surface tension data. Several studies assumed that a gemini surfactant
ion binds one counterion at a concentration below the cmc [1–4]. This assumption was
made because earlier electrical conductivity studies of solutions of alkanediyl-α,ω-
bis(trimethylammonium halide) bolaform electrolytes, which are very similar to the much
investigated m-s-m gemini surfactants, suggested that one counterion was bound to the
bolaform ion [5]. The value n=2 was thus used in the analysis of the surface tension data
for m-s-m surfactants and other dicationic gemini surfactants [1–4] even though there
was, at that time, no result that provided evidence of ion pairing in submicellar solutions
of these surfactants. Other workers used the value n =3 in analyzing surface tension data
or reported two sets of values of a, calculated using the values n=2 and n=3 [6–13].
The knowledge of the value of a is important for evaluating the performances of a
surfactant. Unfortunately, it is clear from the above that the value of a obtained from Eq.
(1) directly depends on the value used for n. Li et al. [14] used the neutron reflectivity
technique in an attempt to solve this problem. Indeed, this technique permits a direct
measurement of the surface ex-cess Γ. The comparison of the Γ values so obtained to
those of (dγ/d ln C)/ RT from surface tension data permits a determination of the value of
n at any surfactant concentration. It is important to recall that the value of n gives direct
information on whether ion pairing occurs in the solution from which adsorption takes
place. The comparison of the values of Γ and of (dγ/d ln C)/RT for 12-s-12 surfactants led
to the following tentative conclusions [14]. In submicellar solutions of 12–3–12, 12–4–
12, and 12–12–12, the value of n was found equal to 2, indicating that the dimeric
surfactant ion binds one bromide ion. The value n=3 was found for 12-pxylyl-12
(pxylyl= para-phenylenedimethylene), indicating a complete ionization of this surfactant
in submicellar solutions. The case of 12–6–12 was more complex as the value of n
decreased from 3 at low C (full ionization) to 2 at higher C, close to the cmc (ion pairing)
[14].
Ion pairing in solutions of 12-s-12 surfactants was also investigated by means of
electrical conductivity measurement [15]. Indeed, the electrical conductivity is an
extremely precise technique that is very sensitive to ion pairing. The results of these
measurements did not support the conclusions reached by Li et al. [14]. In particular, for
Gemini surfactants 94
the surfactant 12–6–12, the increase of the extent of ion pairing with the surfactant
concentration at C<cmc, suggested by the neutron reflectivity results [14], should have
resulted in a considerable curvature of the plot of the specific conductivity versus C
toward the C axis at C<cmc. However, the experimental plot was linear in this range (see
Fig. 1). In addition, the values of the equivalent conductivity calculated from the values
of the conductivity were consistent with fully ionized 12-s-12 surfactants at C<cmc [15].
At this stage, we note that Eastoe et al. [16] pointed out that differences between Γ values
from neutron reflectivity and surface tension data might occur due to effects other than a
partial ionization.
Nevertheless, ion pairing does occur in solutions of m-s-m surfactants with shorter
alkyl chains, i.e., with higher cmc values. For instance, the conductivity versus
concentration plots for the surfactants 8–3–8 and 8–6–8 showed two breaks (see Fig. 2)
[17]. In an early study, the cmc of these surfactants was taken as the concentration
corresponding to the first break [18]. Later, Frindi et al. [17] used pyrene fluorescence
probing and ultrasonic absorption to show that, in fact, micelles are present only at
concentrations above that corresponding to the second break, thus identifying this
concentration to the cmc. The first break is most likely associated to the surfactant
concentration where ion-pair formation becomes significant enough to show by a
curvature of the conductivity plot toward the concentration axis, as is expected from this
effect. The conductivity versus concentration plots for the surfactants 10–3–10 and 10–8–
10 also showed two breaks [15]. The change of slope at the first break was less
pronounced that with solutions of 8-s-8
aggregation were occurring with 8-oxylyl-8 it would do so even more strongly with the
12-pxylyl-12 surfactant. However, electrical conductivity measurements revealed no
premicellar aggregation for this surfactant [15].
In conclusion, ion pairing occurs in solutions of m-s-m gemini surfactants with short
alkyl chains (m≤10), which are characterized by high cmc values. Other types of gemini
surfactants should be investigated from this point of view in order to assess the degree of
generality of this conclusion. Also, additional studies are required to explain the different
conclusions reached in neutron reflectivity and electrical conductivity studies.
III.
PREMICELLAR AGGREGATION IN SOLUTIONS OF GEMINI
SURFACTANTS
A.
Parameters of Premicellization
Premicellization appears to be a rather general effect in gemini surfactant solutions. It has
been observed with surfactants with a spacer that is either rigid or flexible [6,11,15,22–
25].
The electrical conductivity has been a very convenient method for detecting
premicellization [15,26]. This effect is shown by a positive curvature of the specific
conductivity versus C plot below the cmc and also by a maximum in the plot of the molar
conductivity versus C½ (see Fig. 5 for the 16–8–16 surfactant). Both effects arise because
the molar conductivity of a small aggregate of surfactant ions (whether monomeric or
gemini) is larger than the sum of the molar conductivities of the surfactant ions
constituting it [15,26]. Pinazo et al. [26] have dealt with the conductivity of solutions of
gemini surfactants quantitatively and demonstrated the existence of a maxi
State of gemini surfactants in solution at concentrations below the cmc 97
B.
Size of the Premicellar Aggregates
Rosen et al. [22,23] attempted to determine the equilibrium constants for the formation of
premicellar aggregates of gemini surfactants from surface tension data. Their calculations
assumed the formation of a single type of premicellar aggregate (i.e., a dimer, a trimer, or
a tetramer), having very little surface activity. The values of the equilibrium constants for
the oligomers considered in the calculations were all fairly close. The authors concluded
that a series of premicellar aggregates are formed in the submicellar range of
concentration, in solutions of several gemini surfactants.
Pinazo et al. [26] have tried to account quantitatively for the variation of the
equivalent conductivity with the surfactant concentration in the submicellar range for a
gemini surfactant derived from arginine and to estimate the aggregation number of the
premicellar aggregates. Figure 6 shows results of model calculations that suggest that the
premicellar aggregates must have an aggregation number smaller than 5.
Mathias et al. [28] have determined the aggregation number NPA of the premicellar
aggregates formed in submicellar solutions of several gemini surfactants by means of
time-resolved fluorescence quenching. In all instances, the measurements yielded values
of NPA equal to or close to 2. The use of the time-resolved fluorescence quenching
method for such measurements has been criticized [29]. Indeed, it has been pointed that
the method does not apply at the very low concentration at which the measurements were
Gemini surfactants 102
performed in Ref. 28. In addition, the method does not permit measurements of such low
values of the aggregation number [29]. The state of aggregation of long-chain gemini
surfactants in the premicellar range can probably be best investigated by methods that are
sensitive to the number of species present in the solution (osmometry, for instance) or by
electrical conductivity.
C.
Miscellaneous Studies
Results obtained in a microcalorimetric study of 12-s-12 surfactants suggested the
possible contribution of premicellar aggregation to the measured thermal effect [30]. This
possibility was not confirmed in a later study of 12-s-12 surfactants [31]. Later,
premicellization was proposed to explain the peculiar behavior of the dissymmetric
gemini surfactant 18–6–8 [32] in a microcalorimetric study.
Esumi et al. [33] synthesized the trimeric surfactant 12–2–12–2–12 and attempted to
determine its cmc by electrical conductivity measurements. The conductivity increased
continuously with the concentration without showing the break that is usually identified
with the cmc. The cmc showed in the surface tension versus concentration plot, although
the break in the curve corresponding to the cmc did not show as clearly as is usually the
case. Premicellar aggregation may be responsible for this behavior.
Pareira et al. [34] reported the presence of two breaks in the plots of the specific
conductivity versus concentration for many gemini surfactants structurally very similar to
the m-s-m surfactants but where the polymethylene spacer is replaced by a short
hydrophilic poly(ethylene oxide) group of chemical structure CH2CH2(OCH2CH2)x. The
surface tension versus concentation plots showed only one break that coincided with one
or the other of the breaks in the conductivity plots. The authors presented no explanation
for the presence of the break that did not show in surface tension measurements.
Premicellar aggregation may again be responsible for the low concentration breaks.
Indeed, all surfactants investigated had an alkyl chain carbon number m≥14. In addition,
Pinazo et al. [26] showed that the break in the surface tension versus concentration plot
may correspond to the concentration where premicellization becomes significant and not
to the cmc. This depends on the size of the premicellar aggregates and on their surface
activity.
An abnormal variation of log cmc with m (presence of a minimum) was seen in a
study of di(sodium sulfonate) gemini surfactants, at m>12 [35]. The authors did not
comment on this effect, which is most likely due to premicellization.
Finally, it is noteworthy that although gemini surfactants give rise to premicellar
aggregates more readily than the corresponding monomeric surfactants, premicellization
appears to occur more readily with trimeric surfactant than with gemini surfactants. Thus,
the cmc of a series of trimeric surfactants (with three alkyl chains and three sulfonate
head groups) was found to increase with the alkyl chain length already at m=10. This
unusual effect was explained in terms of premicellar association [36,37].
State of gemini surfactants in solution at concentrations below the cmc 103
IV.
INTRAMOLECULAR INTERACTION BETWEEN THE ALKYL
CHAINS OF A GEMINI SURFACTANT
V.
CONFORMATIONAL ASPECTS OF GEMINI SURFACTANTS
Coiling of the surfactant alkyl chain over itself has been suggested as a possible
explanation of the behavior of gemini surfactants at C<cmc [6]. However, this effect is
likely to be small. Indeed, thermodynamic studies have shown that the free energy of
transfer of an alkane from the gas phase or from an organic solvent to water varies
linearly up to a value of the carbon number m=20 [43]. The occurrence of self-coiling at
some value of m should have been shown by a departure from linearity.
Gemini surfactants 104
The most interesting conformational aspects of gemini surfactants involve the spacer.
Thus, the spacer conformation is very important for explaining the variation of the
surface area a occupied by one gemini surfactant at the air-water interface [44,45]. The
variation of a with the spacer carbon number for 12-s-12 surfactants [7] and x-ray studies
of the mesophases formed by these surfactants [46] suggest that the spacer remains at the
air-water interface with a nearly fully extended conformation at least up to s=8. For larger
values of s, the spacer is too hydrophobic to remain in contact with water. The spacer
then adopts a folded or looped conformation and is located at the air side of the interface,
as in the case of a bolaform surfactant [47].
A change of conformation of the gemini surfactant molecule must occur upon
micellization in the case of m-s-m surfactants, at least for short spacers. This can be
simply understood as follows. In m-s-m surfactants, the fully extended spacer has a length
close to ls (nm)=0.126(s+1) at small values of s. Thus, the spacer length is smaller than
the diameter of an alkyl chain (about 0.52 nm) for s<4. Steric hindrance would then
prevent the two alkyl
VI.
CONCLUSIONS
REFERENCES
Gemini surfactants 106
1. Devinsky, F.; Masarova, L.; Lacko, I. J. Colloid Interface Sci. 1985, 105, 235.
2. Devinsky, F.; Lacko, I.; Bittererova, F.; Tomeckova, L. J. Colloid Interface Sci. 1986, 114, 314
and references therein.
3. Devinsky, F.; Lacko, I. Tensides Detergents 1990, 27, 344.
4. Pinazo, A.; Diz, M.; Solans, C.; Pes, M.A.; Era, P.; Infante, M.R. J. Am. Oil Chem. Soc. 1993,
70, 37.
5. Fuoss, R.A.; Chu, V.F. J. Am. Chem. Soc. 1951, 73, 949.
6. Menger, F.M.; Littau, C.A. J. Am. Chem. Soc. 1993, 115, 10083.
7. Alami, E.; Beinert, G.; Marie, P.; Zana, R. Langmuir 1993, 9, 1465.
8. Pinazo, A.; Infante, M.R.; Chang, C.-H.; Franses, E.I. Colloids Surfaces A 1994, 87, 117.
9. Espert, A.; v. Klitzing, R.; Poulin, P.; Colin, A.; Zana, R.; Langevin, D. Langmuir 1998, 14,
4251.
10. Dreja, M.; Pyckhout-Hintzen, W.; Mays, H.; Tiecke, B. Langmuir 1999, 15, 391.
11. Menger, F.M.; Keiper, J.S.; Azov, V. Langmuir 2000, 16, 2062.
12. Takemura, T.; Shiina, M.; Izumi, M.; Nakamura, K.; Miyazaki, M.; Torigoe, K.; Esumi, K.
Langmuir 1999, 15, 646.
13. Wettig, S.D.; Verrall, R.E. J. Colloid Interface Sci. 2001, 235, 310.
14. Li, Z.X.; Dong, C.C.; Thomas, R.K. Langmuir 1999, 15, 4392.
15. Zana, R. J. Colloid Interface Sci. 2002, 246, 182.
16. Eastoe, J.; Nave, S.; Downer, A.; Paul, A.; Rankin, A.; Tribe, K.; Penfold, J. Langmuir 2000,
16, 4511.
17. Frindi, M.; Michels, B.; Levy, H.; Zana, R. Langmuir 1994, 10, 1140.
18. Zana, R.; Benrraou, M.; Rueff, R. Langmuir 1991, 7, 1072.
19. Hattori, N.; Hirata, H.; Okabayashi, H.; O’Connor, C.J. Colloid Polym. Sci. 1999, 277, 361.
20. Zana, R. Langmuir 1996, 12, 1208.
21. Zana, R.; Levy, H. Colloids Surfaces A 1997, 117, 229.
22. Song, L.; Rosen, M.J. Langmuir 1996, 12, 1149.
23. Rosen, M.J.; Liu, L. J. Am. Oil Chem. Soc. 1996, 73, 885.
24. Rosen, M.J.; Mathias, J.H.; Davenport, L. Langmuir 1999, 15, 7340.
25. Castro, M.J.; Kovensky, J.; Cirelli, A.F. Langmuir 2002, 18, 2477.
26. Pinazo, A.; Wen, X.; Pérez, L.; Infante, M.R.; Franses, E.I. Langmuir 1999, 15, 3134.
27. Ralston, A.W.; Eggerberger, D.N.; Du Brow, P.L. J. Am. Chem. Soc. 1948, 70, 977.
28. Mathias, J.H.; Rosen, M.J.; Davenport, L. Langmuir 2001, 17, 6148.
29. Zana, R. Langmuir 2002, 18, 7759.
30. Grosmaire, L.; Chorro, M.; Chorro, C.; Partyka, S.; Zana, R. Prog. Colloid Polym. Sci. 2000,
115, 31.
31. Grosmaire, L.; Chorro, M.; Chorro, C; Partyka, S.; Zana, R. J. Colloid Interface Sci. 2002, 246,
175.
32. Bai, G.; Wang, J.; Wang, Y.; Yan, H.; Thomas, R.K. J. Phys. Chem. B 2002, 106, 6614.
33. Esumi, K.; Taguma, K.; Koide, Y. Langmuir 1996, 12, 4039.
34. Parreira, H.C.; Lukenbach, E.R.; Lindemann, M.K. J. Am. Oil Chem. Soc. 1979, 56, 1015.
35. Zhu, Y.-P.; Masuyama, A.; Nagata, T.; Okahara, M. J. Jpn. Oil Chem. Soc. 1991,40, 473.
36. Masuyama, A.; Yokota, M.; Zhu, Y.-P.; Kida, T.; Nakatsuji, Y. J. Chem. Soc. Chem. Commun.
1994, 1435.
37. Sumida, Y.; Oki, T.; Masuyama, A.; Maekawa, H.; Nishiura, M.; Kida, T.; Nakatsuji, Y.; Ikeda,
I.; Nojima, M. Langmuir 1998, 14, 1750.
38. Rozycka-Roszak, B.; Fisicaro, E.; Ghiozzi, A. J. Colloid Interface Sci. 1996,184, 209.
39. Bai, G.; Yan, J.; Li, Z.; Thomas, R.K. J. Phys. Chem. B 2001, 105, 3105.
40. Bai, G.; Yan, H.; Thomas, R.K. Langmuir 2001, 17, 4501.
41. In, M.; Bec, V.; Aguerre-Chariol, O.; Zana, R. Langmuir 2000, 16, 141.
42. Hattori, N.; Yoshino, A.; Okabayashi, H.; O’Connor, C.J. J. Phys. Chem. B 1998, 102, 8965.
43. Smith, R.; Tanford, C. Proc. Natl Acad. Sci. USA 1973, 70, 289.
State of gemini surfactants in solution at concentrations below the cmc 107
I.
INTRODUCTION
This chapter covers the properties of gemini surfactants in water. Even though it deals
with the onset of micelle formation and thermodynamics of micelle formation, these
properties are not directly related to the presence of micelles. They are measured just
when the first micelles form. The interaction between gemini surfactants and water-
soluble polymers is also considered in this chapter because the interaction takes place at
low surfactant concentration below the cmc (critical micellization concentration) in the
absence of free micelles of gemini surfactant. These aspects of gemini surfactants are not
directly related to the properties of the micelles of gemini surfactants reviewed in Chapter
7.
The chapter is organized as follows. Because surfactants are useful and can be utilized
only if they are sufficiently soluble in water, this chapter starts with a section which
reviews the solubility of gemini surfactants in water with the associated problems of
Krafft temperature for ionic gemini and cloud temperature for nonionic gemini. Section
III reviews the effects of several parameters—in particular the length and nature of the
alkyl chain and spacer—on the cmc of gemini surfactants. Section IV deals with the
thermodynamics of micellization of gemini surfactants (free energy, enthalpy, entropy,
and volume of micelle formation). The last section reviews the interaction between
gemini surfactants and water-soluble polymers.
Throughout this chapter, the properties of gemini surfactants are compared to those of
the corresponding monomeric surfactants, whenever possible. Many of the studies that
are reviewed in this chapter deal with cationic gemini surfactants of the alkanediyl-α,ω-
bis(alkyldimethylammonium bro-mide) type. These surfactants will be denoted by the
usual abbreviation m-s-m, where m and s are the carbon numbers of the alkyl and
alkanediyl groups, respectively (see chapter 1).
Gemini (dimeric) surfactants in water 109
II.
SOLUBILITY OF GEMINI SURFACTANTS: KRAFFT
TEMPERATURE, CLOUD TEMPERATURE
The solubility of anionic and cationic dimeric surfactants is considered in the first two
subsections and data concerning their Krafft temperature TK are presented. A third
subsection discusses the few results available for the solubility of nonionic gemini
surfactants and their cloud temperature Tc. At the outset, it must be pointed out that the
literature reports rather little solubility data for gemini surfactants. Indeed, it is generally
accepted that the solubility of ionic surfactants is high at T>K, whereas that of nonionic
surfactants is high at T<Tc. In this chapter, the Krafft temperature is the temperature
above which a mixture of solid surfactant and water (at a surfactant content say around 1
wt %) turns into a solution. The Krafft temperature is slightly higher than the Krafft point
that is defined as the temperature at which the cmc versus temperature plot intercepts the
solubility versus temperature plot.
A.
Anionic Gemini Surfactants
A very large number of anionic gemini surfactants have been synthesized and
investigated, especially by the group of Okahara and Ikeda in Osaka (see Chapter 2).
These surfactants have sodium sulfate, sulfonate, phosphate, or carboxylate head groups.
Different types of spacers group, mostly of the heteroatomic type, containing oxygen,
nitrogen, or sulfur atoms were used [1–11]. For instance, the properties of two series of
gemini surfactants [CmH2m+1OCH2CH(OZ)CH2O]2Y, with m=8 or 10, Z=Na+SO3−
(sulfate surfactants) or Na+(CH2)3SO3− (sulfonate surfactants), and Y=(CH2)2, (CH2)4,
(CH2CH2O)2(CH2)2, p-phenylene, or o-phenylene have been reported [1,2]. These
disodium surfactants were all readily soluble in water, with Krafft temperature TK<0°C.
Similarly low Krafft temperatures were reported in many other studies [3–11]. The low
Krafft temperature of these surfactants was attributed to the low value of m, below 12 in
most cases, and/ or the hydrophilic nature of the spacer group. Nevertheless, three studies
[4,6,10] involved surfactants with longer alkyl chains, with m up to 16. The Krafft
temperature of some of these surfactants was above 0°C [10]. It increased linearly with m
and decreased rapidly when the number x of ethylene oxide units in the spacer was
increased. It was higher than for the corresponding monomeric surfactants.
The surfactants studied in Refs. 1–11 are highly soluble in water at temperature above
TK.
B.
Cationic Gemini Surfactants
A quick look at the reported values of the solubility and TK of cationic gemini surfactants
would tend to suggest that their solubility is lower and their Krafft temperature higher
than for anionic gemini surfactants. However, a closer examination of the reported data
Gemini surfactants 110
shows that this has to do more with the structure of the investigated cationic surfactants
than with the sign of the charge of the head group. Indeed, many of the investigated
cationic gemini surfactants have a hydrophobic spacer and long alkyl chains.
Many of the synthesized surfactants are made up of two alkyldimethyl-ammonium
halide moieties connected by a spacer group at the level of the charged nitrogen atoms
[i.e., the following structure: CmH2m+1(CH3)2N+-spacer-N+(CH3)2CmH2m+1,2X−
abbreviated m-spacer-m] [12–21]. Some of the TK values reported for these surfactants
are compiled in Table 1. The values of the solubility were in most instances not given. A
comparison of the listed values of TK reveals the following trends.
1. The substitution of Cl− by Br− increases TK (entries 1 and 2; entries 3 and 4). A similar
effect was noted when substituting the bromide anion of 12–8–12 or 12–12–12 by a
methylorange anion [20].
2. As for anionic gemini surfactants [10], TK increases linearly with the alkyl chain
carbon number m (i.e. as the cmc decreases) (entries 2 and 4; entries 12 and 13; entries
16 and 17). This effect is illustrated in Fig. 1 for the surfactants m-
CH2CH2(OCH2CH2)x−m with a short poly(ethylene oxide) hydrophilic spacer [16].
Recall that for conventional surfactants the following empirical relationship holds:
(1)
where k is a constant [23]. Because In cmc varies linearly with m (see Section
III.A.1), the linear plots in Fig. 1 indicate that this relationship also holds for
dimeric surfactants, whether anionic or cationic.
3. TK is increased and the solubility above TK is substantially decreased when the linear
surfactant alkyl chains are replaced by cyclic chains [24]. Thus, the values of TK for
12–3–12 and c12–3–c12 (where c12 stands for cyclododecyl) are respectively 12.7°C
[13] and 31.9°C [24]. The difference is even larger between 12–6–12 and c12–6–c12
(below 0°C and 38.9°C, respectively). The solubility of 12–3–12 is very large at T>TK
TABLE 1 Krafft Temperature and cmc of the
Cationic Dimeric Surfactants
X−,CmH2m+1(CH3)2N+-spacer-+N(CH3)2CmH2m+1,X−
Entry m X− Spacer TKa(°C) cmca(mM)
1 12 Cl− (CH2)3 <0 [12] 0.98 [12]
−
2 12 Br (CH2)3 12.7 [13] 0.81 [21]
−
3 16 Cl (CH2)3 <0 [12] 0.015 [12]
−
4 16 Br (CH2)3 42 [13] 0.0255 [21]
−
5 12 Br (CH2)4 10.6 [13] 1.13 [21]
6 12 Br− (CH2)5 <1 [13] 1.09 [21]
−
7 12 Br (CH2)6 <0 [13] 1.03 [21]
Gemini (dimeric) surfactants in water 111
C.
Nonionic and Zwitterionic Gemini Surfactants
There are relatively few studies of the solubility of nonionic and zwitterionic gemini
surfactants. Depending on the nature of the nonionic surfactant, either a lack of solubility
corresponding to the existence of a Krafft temperature or clouding and separation in two
phases (existence of a cloud temperature) have been reported.
Starting with gemini surfactants with sugar head groups, surfactants with the
disaccharide trehalose as spacer and neutral NH head groups were found to be insoluble
in water [18]. Their Krafft temperature was not reported. The Krafft temperature for the
gemini surfactants with open sugar cycle head groups
[CmH2m+1C(O)NCH2(CHOH)4CH2OH]2(CH2)s was reported to increase substantially with
s [31]. It is noteworthy that the values of TK for the surfactants with m=13 were lower
than for the corresponding bolaform surfactants with m=0 [31]. The Krafft temperature of
the diglucamide surfactants [CmH2m+1]2C[CH2NHCO(CHOH)4CH2OH]2 increased with m
from below 5°C for m=5 and 6 to 35°C for m=7 and 8 [32]. The surfactant with m=7
showed clouding and phase separation at low concentration.
The investigation of the nonionic surfactants [C11H23C(O)NCH2CH2
(OCH2CH2)xOH)]2(CH2)2 revealed that the homolog with x=2 was not soluble in water
[33]. The homolog with x=5 showed clouding and phase separation, with cloud
temperatures between 48 °C and 55°C in the concentration range between 0.1 and 3 wt%
[33]. The corresponding monomers with x=2 and 5 had cloud temperatures above 36°C
and 85°C, respectively. Thus, the dimerization of these nonionic surfactants results in an
unfavorable decrease of the cloud temperature. It is noteworthy that the cloud
temperature Tc of nonionic surfactants, both gemini and conventional, with ethoxylated
head groups goes through a minimum as the surfactant concentration is increased.
Clouding was also observed with nonionic asymmetric gemini surfactants having two
different head groups: a hydroxyl and a methyl-capped poly (ethylene oxide) [34]. The Tc
versus concentration curve also showed a minimum, as for the nonionic gemini
surfactants in Ref. 33 and for conventional nonionic surfactants.
The zwitterionic gemini surfactants CmH2m+1O(O)P(O−)OCH2CH2N+ (CH3)2Cm′H2m′+1
showed limited solubility in water already for m=8 and m′=10 [35].
In conclusion of this section, it appears that the “dimerization” of a conventional
nonionic surfactant results in a gemini surfactant with a lower cloud temperature. For
ionic surfactants, “dimerization” may result in either an increase or a decrease of the
Krafft temperature. In all instances, the values of both TK and Tc for gemini surfactants
with dodecyl chains remain in a range that allows most uses of such surfactants.
III.
CRITICAL MICELLIZATION CONCENTRATION
Practically all articles on gemini surfactants report cmc values. The cmc of gemini
surfactants have been measured using the same methods as for conventional surfactants:
electrical conductivity (for ionic gemini surfactants), surface tension (both static and
Gemini (dimeric) surfactants in water 115
dynamic), dye solubilization, and pyrene intensity ratio I1/I3, to cite but a few. This
section does not present extensive compilations of cmc values. It focuses on the
parameters that determine the cmc and on the analogies and differences between gemini
and conventional surfactants. The effects of the surfactant alkyl chain and of the spacer
are considered successively. This is followed by a review of the cmc of asymmetric
gemini and of oligomeric surfactants. A last subsection compares the cmc of gemini and
of the corresponding monomeric surfactants.
A.
Effect of the Surfactant Alkyl Chain
(2)
Reported values of B and values determined from the reported cmc data, using the linear
part of the plot where premicellization is not present, are listed in Table 2. For the
cationic dimeric surfactants, the values of B are all around 0.43 ± 0.03, except entry 6,
which concerns surfactants investigated at 50°C in water +0.01 M NaCl. The presence of
salt increases the value of B [17,41,42]. The deviation from the average value of B is
Gemini surfactants 116
close to the error in determining this value from the variation of log cmc with m. The
listed results show that B varies little when substituting a bromide by a chloride
counterion (entries 2 and 3), increasing the length of the spacer (entries 1–5), substituting
a hydrophobic spacer by a hydrophilic one (entries 3 and 10), and crowding the charged
nitrogen head group by replacing the methyl groups by bulkier tetramethylene or
pentamethylene groups (compare entry 12 to entries 13 and 14). The B value is also little
affected when replacing the alkyl chains by acyloxymethyl or acyloxyethyl chains
(entries 15 and 16). The only significant
variation of B is seen when going from cationic to anionic dimeric surfactants (entries 17
and 18).
A thermodynamic model of the micellization of gemini surfactants correctly
accounted for the linear decrease of In cmc upon increasing m [45]. Monte Carlo
simulations [46,47] predicted an increase of the cmc with m for ionic dimeric surfactants,
a result that appears to contradict the experimentally observed decrease of cmc upon
increasing m, up to m=16 for many dimeric surfactant series (see above). However, in
these simulations, m represents a number of hydrophobic units and each of these units
may correspond to three or four methylene groups in the surfactant alkyl chain. Thus, a
direct comparison is not really possible at this stage and must wait for additional
simulations [47].
Gemini surfactants 118
2.
Branching of the Alkyl Chain
Branching of the dimeric surfactant alkyl chain results in larger values of the cmc, as for
conventional surfactants [43]. For instance, in Table 2 the values of the cmc of surfactant
15 with m=8 are respectively equal to 8.8, 24, 12, and 13 mM for the n-octyl, iso-octyl,
sec-octyl, and 2-ethylhexyl homologs [43]. This led the authors to determine “effective”
values of m by comparing the cmc values of surfactants with branched alkyl chains and
of chemically identical surfactants with linear alkyl chains.
3.
Nature of the Alkyl Chain
Substitution of alkyl chains by CmH2m+1OC(O)CH2 or CmH2m+1OC(O) CH2CH2 chains
results in a relatively small change of cmc of gemini surfactants. This can be seen by
comparing the values of the cmc of surfactants m-2-m and m1–2-m1 (entries 1 and 15 in
Table 2). For instance, replacing the two C12H25 chains of 12–2–12 by C12H25OC(O)CH2
chains lowered the cmc from 0.85 to 0.2 mM [43]. The difference is mainly due to the
presence of the additional OC(O)CH2 groups in the m1-2-m1 surfactants.
The substitution of CmH2m+1 alkyl chains by chains that are largely fluorinated results
in a large decrease of the cmc. For conventional surfactants, the comparison of the cmc
values of surfactants with hydrogenated and fluorinated alkyl chains revealed that one
CF2 is equivalent to about 1.5 CH2 [48]. This result reflects the fact that the free energy
of transfer from water to the micelle core is of about 1.6 kT/mol CF2 and 1.1 kT/mol CH2
[48]. The same rule appears to apply to dimeric surfactants. For instance, the cmc of 16–
2–16 is 21 µM [21], whereas that of the identical surfactant where the hexadecyl chains
are replaced by C8F17C4H9 chains is 30 µM [49]. Thus, a C8F17 chain is equivalent to a
C12H25 chain.
Dimeric surfactants with hydrophobic moieties containing aromatic groups have been
synthesized [39,50]. The cmc of the surfactants [Br−, ArO (CH2)4N+(CH3)2]2(CH2)6 with
Ar=p-tertiobutylphenyl and 2-naphthyl were found to be close to that for the surfactant
10–6–10 investigated by the same authors [50]. Thus, the p-tertiobutylphenyloxy and 2-
naphthyloxy groups are equivalent to an n-hexyl group as far as the value of the cmc is
concerned.
B.
Effect of the Spacer
1.
Nature of the Spacer
The cmc of many cationic gemini surfactants with a dodecyl chain (m=12) are listed in
Table 1. The value of the cmc is seen to depend remarkably little on the nature of the
spacer, whether hydrophobic (entries 1–11 and 20–22) or hydrophilic (entries 12–19),
flexible (entries 1–9 and 12–18), or rigid (entries 10, 11, and 20–22). Indeed, the values
Gemini (dimeric) surfactants in water 119
of the cmc are all within a factor 2 or so around the value 1 mM. The only exception is
surfactant 19, which contains an extremely bulky, rigid, and hydrophilic disaccharide
spacer and the cmc of which is about three times larger than the average value of 1 mM.
An increased rigidity of the spacer group (as in entries 10, 11, and 19) apparently results
in a slight increase of cmc (i.e., in a more difficult formation of micelles). Note that
surfactants 21 and 22 have rather bulky hydrophobic head groups. Indeed, a
tetramethylene (entry 21) or pentamethylene (entry 22) group bonded to the charged
nitrogen atom by its two ends has replaced the two methyl groups bonded to each
nitrogen atom. Nevertheless, the overall effect of the nature of the spacer is small as
compared to that of the alkyl chain length. The ratio between the lowest and largest cmc
values for the surfactants with m=12 in Table 1 is about 4–5. This ratio corresponds to a
variation by less than two methylene groups in the surfactant alkyl chain. The literature
reports one example of a dimeric cationic surfactant with m= 12 and a cmc around 0.2
mM, which is well below the cmc values for the surfactants in Table 1. It concerns a
dimeric surfactant derived from a hexamethylated biphenyl that plays the role of spacer
[51]. Two dodecyloxy groups and two trimethylammonium chloride head groups are
fixed on the biphenyl. The comparison of the cmc of this surfactant to that of 12-s-12
surfactants [21] indicates that the hexamethylated biphenyl is equivalent to a
polymethylene spacer with s close to 14.
The cmc of anionic gemini surfactants also depends little on the nature of the spacer
[1–11], although perhaps somewhat more than for cationic gemini surfactants. For
instance, the cmc of the surfactants [C8H17OCH2CH(O-SO3−Na+)CH2O]2Y, where the
Y≡(CH2)2, CH2(CH2OCH2)2CH2,p-phenyl-ene, and o-phenylene, have been reported to
be 0.60, 0.18, 0.30, and 0.17 mM, respectively. These four spacers encompass differences
in flexibility, hydro- phobicity, and conformation.
2.
Length of the Spacer
The effect of the length of the spacer has been investigated for flexible hydrophobic and
hydrophilic spacers and for both anionic and cationic gemini surfactants
[3,4,9,10,15,16,50,52–61].
Some results are represented in Figures 4 and 5 for m-s-m surfactants and for anionic
and cationic surfactants with a short poly(ethylene oxide) spacer, respectively. The
difference of behavior is striking, with the cmc going through a maximum for the
surfactants with a hydrophobic polymethylene spacer and increasing slowly and
monotonously with the number of ethylene oxide units for the anionic and cationic
surfactants with a hydrophilic spacer. Figure 4 shows that the maximum of cmc for the
m-s-m surfactants with a hydrophobic polymethylene spacer occurs at s=5–6, irrespective
of the value of m. This value of s is also that for which the melting temperature of the m-
s-m surfactants is a maximum [13]. The explanation of the maximum of cmc probably
involves the following three effects:
1. The first effect is a conformational change of the dimeric surfactant molecule in the
molecularly dispersed state as the length of the spacer is increased (see Chapter 5,
Section V).
Gemini surfactants 120
2. The second effect is a change of conformation forced upon the surfactant molecule as
micellization takes place (see Chapter 5, Fig. 7, for instance). Such a change of
conformation has been evidenced by nuclear magnetic resonance (NMR) [62] and has
also been suggested on the basis of small
3. The last effect corresponds to a progressive incorporation of the spacer in the micellar
core because it becomes increasingly hydrophobic as s is increased.
The variation of the cmc with s would be governed by the effects (1) and (2) at low
values of s, typically s<6. Effect (3) would be the dominant one at larger values of s.
Such an explanation of the results in Fig. 4 is supported by the fact that the maximum of
cmc is not observed with gemini surfactants having a hydrophilic spacer (see Figs. 4 and
5).
Nevertheless, it must be pointed out again that the variations of the cmc with the
spacer length are small, whichever the nature of the spacer. They are much smaller than
with the length of the alkyl chain. This fact is important in the uses and applications of
gemini surfactants because it indicates that the properties (chemical, physical, and
biological) of a gemini surfactant can be much modified by acting on the nature and
length of the spacer, without affecting the cmc substantially.
C.
Asymmetric Dimeric Surfactants
Asymmetric dimeric surfactants (also called heterogemini or heterodimeric surfactants)
are made up of two different amphiphilic moieties, still connected at the level of, or close
to, the head groups. The amphiphilic moieties may differ by the length and/or nature of
the alkyl chains and by the nature of the head groups.
The most investigated asymmetric dimeric surfactants are identical to the m-s-m
surfactants, but the two alkyl chains have different carbon numbers, m and m′ , [63,64].
These surfactants have been denoted m-s-m′. Surfactants with equal m+m′ values have
similar cmc values [63]. For instance, the cmc of 12–2–12, 10–2–14, and 8–2–16 are
respectively 0.96, 0.95, and 0.75 mM. Likewise, the cmc values of 14–2–18 and 16–2–16
are respectively 0.032 and 0.033 mM [63].
An interesting result concerns the effect of the substitution of a dodecyl chain of 12–
2–12 by a C8F17C4H8 group, resulting in an asymmetric gemini surfactant with a
hydrogenated chain and a partially hydrogenated-partially fluorinated chain that can be
denoted 8F4H–2–12 [49]. Its cmc value was found to be 0.2 mM, as compared to 0.89
mM for 12–2–12 and 30 µM for the symmetric surfactant 8F4H–2–8F4H.
Renouf et al. [65] synthesized an asymmetric gemini surfactant with two different
head groups: C10H21CH[CH2(OCH2CH2)xOH](OCH2CH2)2OCH
[CH2O(CH2)3SO3−Na+]C10H21 with x=16–17. Its cmc was found to be about one order of
magnitude lower than for the nonionic conventional surfactant C12H25(OCH2CH2)15OH
that closely corresponds to the nonionic moiety of the gemini surfactant.
Menger et al. [35,66] investigated the fully asymmetric zwitterionic gemini surfactants
CmH2m+1O(O)P(O−)OCH2CH2N+(CH3)2Cm′H2m′+1 (denoted m,m′) that have different head
groups and different alkyl chain lengths. The cmc values for many such surfactants are
Gemini (dimeric) surfactants in water 123
listed in Table 3. Again, the cmc is seen to be mostly dependent on the sum m+m′.
Nevertheless, a small effect of
TABLE 3 cmc of the Asymmetric Zwitterionic
Surfactants CmH2m+1O(O)P(O−)
OCH2CH2N+(CH3)2Cm′H2m′+1
m+m′ m,m′ cmca(mM) m+m′ m,m′ cmca(mM)
16 8,8 1.0 24 10,14 0.0053
18 8,10 0.11 24 14,10 0.0055
18 10,8 0.10 24 12,12 0.0090
20 8,12 0.014 26 8,18 0.0048
20 12,8 0.018 26 18,8 0.0059
20 10,10 0.013 26 10,16 0.0050
22 8,14 0.0061 26 14,12 0.0070
22 14,8 0.0080 26 12,14 0.0043
22 10,12 0.013
a
All cmc values are from Ref. 35.
the asymmetry can be seen by comparing the cmc values for the surfactants m,m′ and
m′,m. Note that all these surfactants have a rather short spacer. The conformational
effects discussed in Section III.B.2 may, therefore, play a role in the observed differences
of cmc between m,m′ and m′,m surfactants. It is noteworthy that starting at m+m′=22, the
cmc becomes nearly independent of the sum m+m′; that is, the ln cmc versus alkyl chain
carbon number plot levels out. This behavior may be the signature of premicellar
aggregation (see Chapter 5, Section III).
Another group of asymmetric gemini surfactants has been synthesized from oleyl
nitrile [34,67,68]. These surfactants can be represented as C8H17
CH[O(CH2CH2O)xCH3]CH(Y)C7H14CN, where the group Y can be OH (nonionic
surfactants) or SO3−Na+ and NH3+Br− (ionic surfactants). For both nonionic [34] and
anionic [68] surfactants, the cmc was found to depend relatively little on the number of
ethylene oxide units x in the head group.
D.
Oligomeric Surfactants
Oligomeric surfactants represent a natural extension of dimeric surfactants. They are
made up of i amphiphilic moieties connected at the level of, or in close vicinity to, the
head groups by i—1 spacer groups. As for dimeric surfactants, the most investigated
trimeric and tetrameric surfactants have been of the quaternary ammonium type. Table 4
summarizes some cmc data for oligomeric surfactants [30,69–72]. The effect of the
spacer length is similar to that described for dimeric surfactants. For instance, the cmc of
Gemini surfactants 124
12-s-12-s-12 surfactants is seen to increase in going from s=2 to s=6. Surfactants 1–4 and
7 all have cmc values between 50 and 280 µM. Those values are one order of magnitude
larger than for surfactants 5 and 6. Such large differences cannot be easily understood. A
possible explanation is that the cmc values for surfac
TABLE 4 cmc of Oligomeric Surfactants
Entry Oligomeric surfactant cmc Ref.
(µM)
1 12–2–12–2–12 80 69
2 12–3–12–3–12 140–160 70
−
3 12–3–12–3–12, 3Cl 330 70
4 12–6–12–6–12 280 71
−
5 12-CH2CHOHCH2-12-CH2CHOHCH2-12,3Cl 9.6 30
6 Identical to entry 5, but the central head group is a neutral nitrogen atom 6.2 30
7 Tri(sodium sulfonate), tridodecyl anionic surfactant with hydrophilic 50 72
spacers
8 12–3–12–4–12–3–12 62 71
tants 5 and 6, obtained from surface tension measurements, represent values of the
concentration where premicellization becomes significant, and not true cmc values. A
similar situation occurred for dimeric surfactants derived from arginine. The cmc for
these surfactants as obtained from surface tension was initially reported to be around 10
µM for m=12 [28,60]. Later, it was realized through fluorescence probing and electrical
conductivity measurements that the true cmc values were much larger, around 400 µM,
and that the previously reported values were those associated with premicellization [73].
The results in Table 4 also show that the substitution of the bromide counterion by a
chloride counterion (surfactants 2 and 3) results in a significant increase of the cmc, as
for conventional and gemini surfactants [70].
The variation of the cmc with the degree of oligomerization for the quaternary
ammonium surfactant oligomers 12-s-12-s-12…is represented in Fig. 6 for the two series
with s=3 and s=6 against the number of dodecyl chains (degree of oligomerization z) in
the surfactant molecule. The cmc has been expressed in equivalent per liter in order to
permit a better appreciation of the decrease of the cmc with increasing number of alkyl
chains. In both instances, the cmc decreases in a somewhat hyperbolic manner in going
from the monomer to the dimer, trimer, and tetramer [71]. The largest decrease of cmc is
obtained in going from the monomer to the dimer. This
Gemini (dimeric) surfactants in water 125
E.
Comparison of the cmc of Gemini Surfactants and of Conventional
Surfactants
In several instances, it has been stated in the literature that the cmc of gemini surfactants
is two or three orders of magnitude smaller than that of conventional surfactants. This
does not appear to be correct. Indeed, one must be very careful when comparing a
conventional and a dimeric surfactant. The former must be the true monomeric surfactant
corresponding to a dimeric surfactant and this was often not the case in the reported
comparisons. In addition, one must be sure that the comparison concerns the value of the
Gemini surfactants 126
cmc of the gemini surfactant and not that of some other concentration—for instance, that
where premicellization or ion pairing become significant (see Chapter 5, Section III).
This possibility was illustrated in the preceding paragraph by the results reported for
dimeric surfactants derived from arginine [28,60,73].
Table 5 lists the cmc values of several pairs of conventional and dimeric surfactants
[21,33,73–75]. It is seen that the ratio cmcmonomer/cmcdimer is around 10 or so for
surfactants with a dodecyl alkyl chain. A ratio around 10 can also be seen in Fig. 5 for the
surfactant series 1 and 2 with dodecyl chains [10]. The ratio is somewhat larger for the
surfactants with a hexadecyl chain, as expected. A surprising result is seen in Table 5 for
the surfactants with glucopyranoside head groups, with a cmc for the dimeric surfactant
that is larger than for the monomeric surfactant [75]. This behavior may be due to the
presence of the hydrophilic succinyl spacer group that connects the two amphiphilic
moieties and also to a premicellization of the dimeric surfactant. Both effects add up,
resulting in a large cmc value for the gemini surfactant.
IV.
THERMODYNAMICS OF MICELLIZATION OF GEMINI
SURFACTANTS
A.
Free Energy of Micellization
The free energy of micellization, , of a surfactant is usually obtained from the value
of the cmc. The first reported values of for many ionic dimeric surfactants
[37,39,43,44] were obtained from the equation
(3)
TABLE 5 Comparison of the cmc Values of
Gemini Surfactants and of the Corresponding
Conventional Surfactants
Conventional surfactant Gemini surfactant
Surfactant cmc Surfactant cmc
(mM) (mEq/L)
C12H25N+(CH3),Br− 16 [21] 12–2–12 1.7 [21]
12–3–12 1.8 [21]
Na+(C10H21CH(OH) 0.72 [Na+,C10H21CHCH2O(CH2)3SO3−]2O 0.19 [74]
CH2O(CH2)3SO3− [74]
C11H23C(O)N(CH3) 0.5 [33] [C11H23C(O)NCH2CH2(EO)5OH]2(CH2)2a 0.04 [33]
CH2CH2(EO)5OHa
Dodecylarginine surfactant 6 [73] Dimer of the dodecylarginine surfactant ~0.8 [73]
Dodecylglucopyranoside 2.3 [75] (Dodecylglucopyranoside)2succinyl 6.8 [75]
Gemini (dimeric) surfactants in water 127
Unfortunately, Eq. (3) holds only for nonionic surfactants. In later studies of cationic
gemini surfactants [76–78], the values of calculated from
(4)
In Eq. (4), α is the micelle ionization degree. This equation holds for conventional ionic
surfactants (one chain connected to a monovalent charged group with a monovalent
counterion) but does not apply to gemini surfactants. The correct equation for gemini
surfactants is [79]
(5)
In Eq. (5), is the free energy of micellization per mole of gemini surfactant. The
cmc is expressed in mole of alkyl chain or in mole fraction of alkyl chain. Note that
because the cmc is low, the mole fraction can be taken as Cchain/55.5. Going from
concentration to mole fraction makes the free energy more negative by −2RT(1.5—α) In
(55.5) ≈ −26 kJ/mol for α ≈ 0.2.
Values of for the 12-s-12 surfactants have been reported in two studies that used the
same set of cmc values but different sets of a values. One study [80] used an ancient set
of a values [21] obtained from conductivity data analyzed by a method [81] that has been
shown to yield overestimated values of α [82]. The other study [83] used a values
obtained from the same conductivity data analyzed using the method proposed by Evans
[84]. The a values thus obtained are more reliable that those obtained by the first method
(see Chapter 7, Section I). Table 6 lists the values of obtained with this new set of values
of a [83]. The listed values differ from those reported in Ref. 83 because the cmc has
been expressed in mole fraction of dodecyl chain rather than in mole of surfactant per
liter. The values of −
TABLE 6 Free Energy, Enthalpy, and Entropy of
Micellization of 12-s-12 Surfactants at 25 °C
Surfactant cmc (mM) αa
are seen to vary relatively little with the spacer carbon number, going through a rather flat
minimum at around s=8. A similar result was reported by Bai et al. [78,80].
The value of has been found to vary little when the spacer is made more
hydrophilic. The reported values of − for the three surfactants 12–4–12, 12-
CH2CHOHCH2CH2–12 and 12-CH2CHOHCHOHCH2-12 are respectively 41.5, 43.2, and
43.7 kJ/mol [85].
The values of for the 16-s-16 surfactant series can be calculated from reported
values of the cmc [21] and assuming that the values of a are the same as for the
corresponding 12-s-12 surfactants. The much lower cmc values of the 16-s-16 surfactants
result in much more negative values of than for 12-s-12 surfactants. For instance,
for 16–3–16, =−92.4 kJ/mol as compared to −66.9 kJ/mol for 12–3–12.
The variation of with the gemini surfactant alkyl chain length has been reported
for the m-6-m series [79]. The plot is linear and yields an increment of free energy of
micellization per methylene group (CH2)= −3.1±0.3 kJ/mol. This value is very close
to that found for the conventional surfactants, alkyltrimethylammonium bromides. The
nonionic diglucamide surfactants [CmH2m+1]2C[CH2NHCO(CHOH)4CH2OH]2 are
characterized by nearly the same value of (CH2) [32].
The free energy of micellization of an oligomeric surfactant can be calculated from
[79]
(6)
In Eq. (6), is expressed in per mole of alkyl chain and z is the surfactant
oligomerization degree. appears to depend only slightly on z. The values of for
the surfactants 12–3–12, 12–3–12–3–12, and 12–3–12–4–12–3–12, calculated from Eq.
(6), were found to be −34.1, −37.1, and −37.1 kJ/mol dodecyl chain, respectively (with
the cmc expressed in mole fraction of dodecyl chain per liter). The calculations assumed
a micelle ionization degree of 0.16 for all three surfactants (see Table 6) and used the
reported cmc values [71]. The value of for DTAB, which can be considered as the
monomer of these oligomeric surfactants, is −36.4 kJ/mol. Similarly close values of
were found for the surfactants 12–6–12, 12–6–12–6–12, and the corresponding
monomeric surfactant 12–3 [71].
It is noteworthy that in systems where premicellization takes place, the apparent cmc
is increased, resulting in less negative values of (see Chapter 5). Indeed, some
water-alkyl chains contacts are eliminated when premicellization occurs and do not
contribute to the free energy of micellization. The free energy of premicellization has
been evaluated from the departure of linearity of the plot of In cmc versus m at large
values of m [40].
Gemini (dimeric) surfactants in water 129
B.
Enthalpy and Entropy of Micellization
Enthalpies of micellization ( ) can be obtained from the temperature dependence of
the cmc. This method is an easy one, as it only involves cmc determinations, but it suffers
from two pitfalls. The first one is that the cmc varies little with temperature and often
goes through a minimum at 20–40 °C. The second one is that the full analysis of the data
requires the temperature derivatives of the micelle ionization degree and aggregation
number. These two facts result in a large uncertainty in the values of so obtained.
Direct calorimetric measurements of have become easy, accurate, and fairly rapid
with the advent of modern isothermal microcalorimeters. This technique has been much
used for measuring the enthalpy of micellization of the gemini surfactants 12-s-12 and of
the corresponding monomeric surfactants 12-s/2 [77,78,80,83,86,87]. The entropy of
micellization can then be obtained from
(7)
C.
Volume of Micellization
Micellization of conventional surfactants always gives rise to a volume increase. The
volume of micellization is also positive for gemini surfactants. It has been directly
Gemini (dimeric) surfactants in water 131
determined from density data [24,77,85,88] and indirectly from ultrasonic relaxation
measurements [38].
As for conventional surfactants, the volume of micellization of gemini surfactants
increases with the alkyl chain length [77,88]. In the case of 12-s-12 surfactants, the
substitution of linear dodecyl groups by cyclic dodecyl groups results in much lower
values of [24].
The effect of the spacer carbon number on for m-s-m surfactants has been
investigated [38,88]. Figure 7 shows the variation of 12-s-12 surfactants with s. The
volume of micellization goes through a broad minimum at s around 5–6.
Conflicting conclusions have been reached in the literature when comparing the
volume of micellization of a dimeric surfactant to that of the corresponding monomeric
surfactant. For the short-chain surfactant 8–6–8, was found to be about twice larger
than for the monomeric surfactant octyltrimethylammonium bromide [38]. The same is
true for 8–3–8 [38,88], 10–3–10 [88] and the surfactant
[Cl−,C10H21OC(O)CH2N+(CH3)2]2CH2CH2 but not for the surfactant
[Cl−,C12H23OC(O)CH2N+(CH3)2]2CH2CH2 [77]. However, the comparison is more
difficult and, in fact, practically impossible to perform for gemini surfactants with m≥12,
owing to their low cmc value. Indeed, the values of the partial molal volumes at
concentrations below 1 mM are extremely inaccurate. This prevents the determination of
.
The volume of micellization has been reported to increase when the spacer is
progressively made more hydrophilic [85].
V.
INTERACTIONS BETWEEN GEMINI SURFACTANT AND
WATER-SOLUBLE POLYMERS
A.
Nonionic Water-Soluble Polymers
The interaction of cyclodextrins and cyclodextrin derivatives with gemini surfactants,
either nonionic [91] or cationic m-6-m [92], has been investigated. Cyclodextrins are
made up of six to eight glucopyranose units and are much used in the pharmaceutical
industry as stabilizing and solubilizing agents. The cmc of the nonionic asymmetric
Gemini surfactants 132
B.
Charged Polymers
Surfynols are nonionic gemini surfactants with an acetylenic spacer. They are
nonfoaming and fast spreading. Their interaction with anionic copolymers of styrene and
ammonium maleate has been investigated by dynamic surface tension measurements, as
this system is a model one for printing inks [98]. The results permitted the determination
Gemini surfactants 134
of the binding isotherms of the surfactants. Binding occurred only at above a certain
value of the surfactant concentration (see Fig. 9), as in the case of conventional
surfactants [90].
Binding isotherms could not be determined for systems containing 12-s-12 surfactants
and oppositely charged polyelectrolytes. Indeed, the interaction is so strong that
precipitation of a solid polymer-surfactant complex occurs as
stronger than with the corresponding monomeric surfactants [99]. The precipitation of a
polymer-surfactant complex does not occur in the presence of a fairly high content of salt
and the investigation of the interaction becomes possible [100,101]. Indeed, the co-ions
from the added salt compete with the dimeric surfactant ions for the polymer binding
sites, thereby reducing the strength of the polymer-surfactant interaction and preventing
the precipitation.
There have been several studies of the interaction of gemini surfactants with DNA. A
first study involved chemically complex gemini surfactants based on tartaric acid [102].
The authors investigated the toxicity of the surfactants and their interaction with the p-
CMV-luciferase plasmid DNA in view of testing the gemini surfactant-DNA complexes
in gene transfection. In another investigation [103], the efficiency of compaction of DNA
by 12-s-12 surfactants was found to be a minimum at around s=6. This value of s is close
to that where several properties of 12-s-12 surfactants go through a maximum or a
minimum (see Sections III.B and IV). The compaction of the DNA molecule is a
necessary requirement for transfection. The last study concerns 12-s-12/DNA complex
monolayers at the air-water interface [104]. The results suggest that the surfactant spacer
in the complex adopts a looped conformation starting at s=6. The spacer conformation
appears to have an important effect on the structure of the monolayer as visualized by
atomic force microscopy.
As can be seen from the above, only fragmentary studies have been reported thus far
on gemini surfactant-polymer systems. More systematic investigations are required in
order to reach a better understanding of these systems. Also, some studies of dimeric
surfactant-oppositely charged polymer complexes in the solid state should be performed.
Some interesting structures may be discovered in the process.
VI.
CONCLUSIONS
This chapter reviewed various aspects of gemini surfactants. The cmc, which is probably
the most important characteristic of a surfactant, is about one order of magnitude lower
for a gemini surfactant than for the corresponding conventional surfactant. It depends
relatively little on the length and nature of the spacer. This makes it possible to confer to
a gemini surfactant the specific chemical or biological properties associated to the spacer
without substantially changing the onset of self-association. Another important result is
the fact that gemini surfactants interact much more strongly with water-soluble polymers
than the corresponding monomeric surfactants. This property may prove useful in certain
uses of polymer-surfactant systems. Nevertheless, the need for additional studies for a
fuller understanding of the properties of gemini surfactants was emphasized in several
instances in this chapter.
Gemini surfactants 136
REFERENCES
1. Okahara, M.; Masuyama, A.; Sumida, Y.; Zhu, Y.-P. J. Jpn. Oil Chem. Soc. 1988, 37, 58.
2. Zhu, Y.-P.; Masuyama, A.; Okahara, M. J. Am. Oil Chem. Soc. 1990, 67, 459.
3. Zhu, Y.-P.; Masuyama, A.; Okahara, M. J. Am. Oil Chem. Soc. 1991, 68, 268.
4. Zhu, Y.-P.; Masuyama, A.; Nagata, T.; Okahara, M. J. Jpn. Oil Chem. Soc. 1991, 40, 473.
5. Zhu, Y.-P.; Masuyama, A.; Kirito, Y.; Okahara, M. J. Am. Oil Chem. Soc. 1991, 68, 539.
6. Masuyama, A.; Hirono, T.; Zhu, Y.-P.; Okahara, M.; Rosen, M.J. J. Jpn. Oil Chem. Soc. 1992,
41, 13.
7. Zhu, Y.-P.; Masuyama, A.; Kirito, Y.; Okahara, M.; Rosen, M.J. J. Am. Oil Chem. Soc. 1992, 69,
626.
8. Zhu, Y.-P.; Masuyama, A.; Nakatsuji, Y.; Okahara, M. J. Jpn. Oil Chem. Soc. 1993, 42, 86.
9. Zhu, Y.-P.; Masuyama, A.; Kobata, Y.; Okahara, M.; Rosen, M.J. J. Colloid Interface Sci 1996,
158, 40.
10. Okono, T.; Egawa, N.; Fujiwara, M.; Fukuda, M. J. Am. Oil Chem. Soc. 1996, 73, 31.
11. Masuyama, A.; Endo, C.; Takeda, S.; Nojima, M.; Ono, D.; Takeda, T. Langmuir 2000, 15,
368.
12. Kim, T.S.; Kida, T.; Nakatsuji, Y.; Hirao, T.; Ikeda, I. J. Am. Oil Chem. Soc. 1996, 73, 907.
13. Zana, R. J. Colloid Interface Sci 2002, 252, 259.
14. Dreja, M.; Tieke, B. Ber. Bunsenges. Phys. Chem. 1998, 102, 1705.
15. Dreja, M.; Pickhout-Hintzen, W.; Mays, H.; Tieke, B. Langmuir 1999, 15, 391.
16. Parreira, H.C.; Lukenbach, E.R.; Lindemann, M.K. J. Am. Oil Chem. Soc. 1979, 56, 1015.
17. Rosen, M.J.; Liu, L. J. Am. Oil Chem. Soc. 1996, 73, 885.
18. Menger, F.M.; Mbadugha, B.N. J. Am. Chem. Soc. 2001, 123, 875.
19. Zhao, J.; Christian, S.D.; Fung, B.M. J. Phys. Chem. B 1998, 102, 7613.
20. Buwalda, R.T.; Engberts, J.B.F.N. Langmuir 2001, 17, 1054.
21. Zana, R.; Benrraou, M.; Rueff, R. Langmuir 1991, 7, 1072.
22. Menger, F.M.; Keiper, J.S.; Azov, V. Langmuir 2000, 16, 2062.
23. Gu, T.; Sjoblom, J. Colloids Surfaces 1992, 64, 39.
24. Jenkins, K.M.; Wettig, S.D.; Verrall, R.E. J. Colloid Interface Sci. 2002, 247, 456.
25. Menger, F.M.; Littau, C.A. J. Am. Chem. Soc. 1993, 115, 10083.
26. Shinoda, K.; Maekawa, M.; Shibata, Y. J. Phys. Chem. 1986, 90, 1228.
27. Gaysinski, M.; Joncheray, L.; Guittard, F.; Cambon, A.; Chang, P. J. Fluor. Chem. 1995, 74,
131.
28. Pérez, L.; Torres, J.L.; Manresa, A.; Solans, C.; Infante, M.R. Langmuir 1996, 12, 5296.
29. Kunieda, H.; Masuda, N.; Tsubone, K. Langmuir 2000, 16, 6438.
30. Kim, T.-S.; Kida, T.; Nakatsuji, Y.; Ikeda, I. Langmuir 1996, 12, 6304.
31. Pestman, J.; Terpstra, K.R.; Stuart, M.C.; van Doren, H.A.; Brisson, A.; Kellogg, R.M.;
Engberts, J.B.F.N. Langmuir 1997, 13, 6857.
32. Eastoe, J.; Rogueda, P.; Howe, A.; Pitt, A.R.; Heenan, R.K. Langmuir 1996, 12, 2701.
33. Paddon-Jones, G.; Regismond, S.; Kwetkat, K.; Zana, R. J. Colloid Interface Sci. 2001, 243,
496.
34. Alami, E.; Holmberg, K. J. Colloid Interface Sci. 2001, 239, 230.
35. Peresypskin, A.V.; Menger, F.M. Org. Lett. 1999, 1, 1347.
36. Zana, R.; Levy, H. Colloids Surfaces A 1997, 117, 229.
37. Devinsky, F.; Lacko, I.; Bittererova, F.; Tomeckova, L. J. Colloid Interface Sci. 1986, 114, 314.
38. Frindi, M.; Michels, B.; Levy, H.; Zana, R. Langmuir 1994, 10, 1140.
39. Devinsky, F.; Lacko, I.; Mlynarcik, D.; Racansky, V.; Krasnec, L. Tenside Detergents 1985, 22,
10.
Gemini (dimeric) surfactants in water 137
82. Sugihara, G.; Nakamura, A.; Nakashima, T.; Araki, Y.; Okano, T.; Fujiwara, M. Colloid Polym.
Sci. 1997, 275, 790.
83. Grosmaire, L.; Chorro, M.; Chorro, C.; Partyka, S.; Zana, R. J. Colloid Interface Sci. 2002, 246,
175.
84. Evans, H.C. J. Chem. Soc. 1956, 579.
85. Wettig, S.D.; Nowak, P.; Verrall, R.N. Langmuir 2002, 18, 5354.
86. Grosmaire, L.; Chorro, M.; Chorro, C.; Partyka, S.; Zana, R. Prog. Colloid Polym. Sci. 2000,
115, 31.
87. Bai, G.; Wang, Y.; Yan, H.; Thomas, R.K. J. Colloid Interface Sci. 2001, 240, 375.
88. Wettig, S.D.; Verrall, R.N. J. Colloid Interface Sci. 2001, 235, 310.
89. Bai, G.; Wang, J.; Wang, Y.; Yan, H.; Thomas, R.K. J. Phys. Chem. B 2002, 106, 6614.
90. Goddard, E.D. Colloids Surfaces 1986, 79, 255; 1986, 79, 301.
91. Abrahmsen-Alami, S.; Alami, E.; Eastoe, J.; Cosgrove, T. J. Colloid Interface Sci. 2002,246,
191.
92. Kralova, K.; Mitterhauszerova, L.; Szejtli, J. Tenside Detergents 1983, 20, 37.
93. Satake, I.; Ikenoue, T.; Takeshita, T.; Hayakawa, K.; Maeda, T. Bull. Chem. Soc. Japan 1985,
55, 2746.
94. Wettig, S.D.; Verrall, R.E. J. Colloid Interface Sci. 2001, 244, 377.
95. Kästner, U.; Zana, R. J. Colloid Interface Sci 1999, 275, 468.
96. Bai, G.; Wang, Y.; Yan, H.; Thomas, R.K.; Kwak, J.C.T. J. Phys. Chem. B 2002, 106, 2153.
97. Esumi, K.; Saika, R.; Miyazaki, M.; Torigoe, K.; Koide, H. Colloids Surfaces A 2000, 166, 115.
98. Krishnan, R.; Sprycha, R. Colloids Surfaces A 1999, 149, 355.
99. Zana, R.; Benrraou, M. J. Colloid Interface Sci. 2000, 226, 286.
100. Pisarcik, M.; Soldan, M.; Bakos, D.; Devinsky, F.; Lacko, I. Colloids Surfaces A 1999, 150,
207.
101. Pisarcik, M.; Imae, T.; Devinsky, F.; Lacko, I.; Bakos, D. J. Colloid Interface Sci. 2000, 225,
207.
102. Buijnsters, P.J.; Garcia, C.L.; Willighagen, E.L.; Sommerdijk, N.A.; Kremer, A.; Camilleri,
P.; Feiters, M.C.; Nolte, R.J.; Zwannenburg, B. Eur. J. Org. Chem. 2002, 8, 1397.
103. Karlsson, L.; van Ejik, M.C.P.; Söderman, O. J. Colloid Interface Sci. 2000, 252, 290.
104. Chen, X.; Wang, J.; Shen, N.; Luo, Y.; Li, Liu, L., M.; Thomas, R.K. Langmuir 2002, 18,
6222.
7
Properties of Micelles and of Micellar
Solutions of Gemini (Dimeric) Surfactants
RAOUL ZANA Institut C.Sadron (CNRS-ULP), Strasbourg, France
I.
INTRODUCTION
This chapter reviews the properties of the micelles of gemini (dimeric) surfactants and of
micellar solutions of these surfactants. The following aspects are successively reviewed:
micelle degree of ionization (Section II), size and polydispersity of the micelles (Section
III), micelle shape and microstructure in solutions of gemini surfactants (Section IV),
micelle micropolarity and microviscosity (Section V), dynamics of gemini surfactant
micelles (Section VI), and solubilization in micellar solutions of gemini surfactants of
compounds sparingly soluble in water (Section VII). A short glossary is given at the end
of this chapter.
II.
DEGREE OF IONIZATION OF IONIC GEMINI SURFACTANT
MICELLES
A.
Methods for the Determination of the Micelle Ionization Degree
Essentially two types of measurement have been used to determine the value of the
micelle ionization degree a of gemini surfactant micelles: electrical conductivity and
small-angle neutron scattering (SANS).
Electrical conductivity measurements are often used for critical micellization
concentration (cmc) determinations. However, the data can also be used for the
determination of α at the cmc. For both conventional and gemini surfactants, a has often
been obtained as the ratio of the slopes of the plot of the specific conductivity K versus
surfactant concentration C above and below the cmc [1–11]; that is,
(1)
The main approximation underlying this expression is that a micellized surfactant ion
whose electrical charge is not compensated by a bound counterion contributes to the
conductivity of the solution the same amount as if it is free [1]. This approximation
Gemini surfactants 140
underestimates the micelle contribution to the conductivity of the solution and results in
overestimated values of α. The error increases rapidly with the value of α [12]. The
method proposed long ago by Evans [13] for the analysis of conductivity data in view of
the determination of the value of a accounts more correctly for the micelle contribution to
the conductivity of the solution, and a at the cmc is obtained by solving the equation
(2)
In Eq. (2), Ax is the equivalent conductivity of the counterion X. When the cmc is low,
the value of Ax at infinite dilution can be used in the calculation of α. The micelle
aggregation number at the cmc, Ncmc, is obtained from the extrapolation of the N versus C
plot to the cmc. A more sophisticated method of analysis of the conductivity data that
accounts for the effect of the ionic strength has been proposed [14]. The values of α and
of the micelle aggregation number N are two adjustable parameters, but the method
assumes a constant value of N at C>cmc. This method has not been applied to the
conductivity data for gemini surfactants.
In SANS experiments, α is obtained from the fitting of a theoretical expressions of the
scattered intensity to the plots of the scattered intensity versus scattering vector for a
surfactant solution at a given concentration [15– 22]. The micelle ionization degree can
therefore be obtained at any surfactant concentration. However, the scattered intensity
depends mostly on the size and shape of the micelles, and the error on the value of α can
be large. The same remark holds for light scattering.
The recently proposed chemical trapping method has been used to determine the
interfacial concentration of counterions at the micelle surface [7]. Although the
relationship between this concentration and α remains to be worked out, these two
quantities should follow the same trends for a series of homologous surfactants.
Potentiometric measurements with counterion-selective electrodes [23] have not yet
been used for the determination of α of gemini surfactant micelles. Also, a recently
proposed method based on the use of electron paramagnetic resonance (EPR) probes has
not yet been applied to gemini surfactants [24].
B.
Review of the Reported Results
Most of the reported results concern the m-s-m surfactants. The variation of a for 10-s-10
surfactants (values from SANS) and 12-s-12 surfactants (values from conductivity
measurements) with the spacer carbon number s are represented in Fig. 1. The values of α
for the 12-s-12 surfactants are larger when calculated from Eq. (1) than from Eq. (2),
particularly at larger values of s when a is large, as expected from the above discussion.
Nevertheless, the same trend is obtained, with an increase of α with s. The value of the
interfacial concentration of counterions for 12-s-12 micelles was found to increase with
the spacer carbon number [7], a result in agreement with the trends in a just discussed. An
increase of α with s was also reported for the micelles of the surfactants 16-s-16 [16,19]
as well as for the micelles of the anionic gemini surfactants 16-s-16,2Na+ that have a
polymethylene spacer and sodium phosphate head groups [20].
Properties of micelles and of micellar solutions of gemini (dimeric) surfactants 141
The comparison of the α values for the 10-s-10 [17,18], 12-s-12 [10], and 16-s-16
[16,19] surfactants series show that α decreases upon increasing surfactant chain length.
This is also seen in the results tabulated in Ref. 8. A
All of the above results refer to gemini surfactants with a hydrophobic polymethylene
spacer. The situation is apparently more complex for the m-EOx-m surfactants with a
short hydrophilic poly(ethylene oxide) spacer EOx=CH2CH2(OCH2CH2)x. For instance,
the ionization degree of 16-EOx-16 micelles is a maximum for x=2 at a concentration of
30 mM [20], but is nearly independent of x at a concentration of 50 mM [21]. On the
contrary, the ionization degree of 12-EOx-12 micelles increases rapidly with x at
concentrations of 60 and 120 mM [15]. No explanation can be given for this difference of
behavior between 12-EOx-12 and 16-EOx-16 surfactants. Nevertheless, for the two series
of surfactants, α increases with the spacer length (i.e., with x) and decreases upon
increasing concentration, as for gemini surfactants with a hydrophobic spacer and for
conventional surfactants.
A direct comparison of the effect of the nature of the spacer for the m-s-m and the m-
EOx-m surfactants on the values of α is not possible because these two series of
surfactants were investigated by means of very different techniques (electrical
conductivity and SANS, respectively). However, the value of a was found to depend little
on the nature of the spacer for the three surfactants 12–4–12, 12-CH2CHOHCH2CH2–12,
and 12-CH2CHOHCHOHCH2– 12 with increasingly hydrophilic spacer [11].
The few available values of α for oligomeric surfactants of the quaternary ammonium
type show little effect of the surfactant degree of oligomerization. Thus, for the bromide
surfactants dodecyltrimethylammonium bromide (DTAB), 12–3–12, 12–3–12–3–12, and
12–3–12–4–12–3–12, α decreases a little in going from DTAB to 12–3–12 and then
remains nearly constant [4,6]. Likewise, nearly the same value of α was found for 12–6–
12, 12–6–12–6–12, and the conventional surfactant 12,3
(dodecylpropyldimethylammonium bromide) that can be considered as the monomer of
12–6–12 and 12–6–12–6–12 [6]. As for conventional surfactants, the chloride oligomers
have a larger micelle ionization degree than the corresponding bromides [4].
III.
SIZE AND POLYDISPERSITY OF GEMINI SURFACTANT
MICELLES
A.
Methods for the Determination of the Micelle Size
Small-angle neutron scattering [15–22,25–27], time-resolved fluorescence quenching
(TRFQ) [4,6,9,11,28–31], and static or dynamic light scattering [32–34] have been used
to determine the size and polydispersity of micelles of dimeric and trimeric surfactants. In
this section, “size” refers to the micelle diameter, the micelle aggregation number N
(number of surfactants per micelle), or the micelle molecular weight. In principle, from
any one of these three quantities one can obtain the other two. SANS and dynamic light
scattering (DLS) measure the micelle diameter [35,36]. SANS is more powerful that DLS
because the measurements can be performed in a wide range of scattering vector and also
because the contrast between the investigated particle and the solvent can be easily
adjusted [35]. Static light scattering measures the micelle molecular weight at the cmc
[36]. TRFQ gives a direct access to the micelle aggregation number in the actual
Properties of micelles and of micellar solutions of gemini (dimeric) surfactants 143
B.
Micelle Aggregation Number: Review of the Reported Results
1.
m-s-m Surfactants
Extensive measurements of micelle aggregation number have been performed for m-s-m
gemini surfactants, that is surfactants with a hydrophobic polymethylene spacer. Figures
2 and 3 show the variations of N with the surfactant concentration for several 10-s-10
[17,18] and 12-s-12 [4,29] surfactants.
In Figs. 2 and 3, all N versus C plots extrapolate to about the same value of N at the
cmc. A similar behavior is observed with the less extensive results
Gemini surfactants 144
to micelle growth. At this stage, it is interesting to note that the N versus s plot for the
surfactants 1–1′-(1-ω-alkanediyl)bispyridinium tetradecanesulfonates also goes through a
minimum at s around 8 [42]. In these surfactants, the counterion is a bolaform ion with a
polymethylene spacer (CH2)s. This behavior has been explained by assuming that the
spacer penetrates into the micelle core at s≥8 [42], just like the spacer group of the m-s-m
surfactants.
The results in Fig. 2 for the 10-s-10 surfactants have been analyzed in terms of the
ladder model for micelle growth [43]. This analysis yielded the values of , the free
energy difference between N0 dimeric surfactants packed in a part of a cylindrical micelle
and in the maximum spherical micelle of aggregation number N0. The variation of
with s is represented in Fig. 5. The values of are all negative. | |is largest at s
=2. It then decreases, goes through a minimum at s=8–10, and increases again as s is
increased. The minimum occurs at a value of s close to that for which the surfactant has
the smallest aggregation number at low concentration, and the micelles are spherical.
This is clearly in relation with the s dependence of the surfactant packing parameter P
discussed earlier. Indeed, the value s=8–10 is close to that for which the optimal surface
area a occupied by one surfactant at the air-water interface is a maximum (see Chapter 4,
Fig. 1) and thus P is a minimum. Spherical micelles are then expected, as experimentally
observed. Models more sophisticated than the ladder model have been used to explain
micelle growth in solutions of conventional and gemini surfactants [44–46].
The replacement of the linear dodecyl chains of 12-s-12 surfactants by cyclododecyl
groups results in much lower values of the aggregation number [9]. Likewise, the
substitution of the bromide counterions by chloride counterions brings about a significant
decrease of aggregation number [4]. Similar results have been reported for conventional
surfactants.
A small decrease of N was reported to occur as the tetramethylene spacer of the 12–4–
12 surfactant was made more hydrophilic as in the 12-CH2CHOH-CH2CH2-12 and 12-
CH2CHOHCHOHCH2-12 surfactants [11].
Gemini surfactants 148
2.
m-EOx-m Surfactants
The above results referred to gemini surfactants with a hydrophobic spacer. An increase
of the value of N with C has also been reported for the surfactants m-EOx-m with m=12
[15] and 16 [20,21]. However, the effect of the spacer length on N is more complex than
for the m-s-m surfactants. The 16-EOx-16 surfactants clearly show a decrease in the
tendency for micelle growth upon
TABLE 1 Values of the Aggregation Number of
Micelles of Gemini Surfactants at 30°C
Entry Surfactant N
+
1 16–6–16,2Na 76
2 16–6–16 67
+
3 16–10–16,2Na 58
4 16–10–16 50
5 16–5–16 124
6 16–EO1–16 193
7 16–5–16 124
8 16–E02–16 119
9 16–8–16 66
10 16–EO3–16 180
11 16–10–16 70
Note: Surfactant concentration: 10 mM for entries 1–4 and 30 mM for entries 5–11.
Source: Ref. 20.
increasing spacer length (i.e., the number x of ethylene oxide units) [20,21]. On the
contrary, the reported results for the 12-EOx-12 surfactants show about the same increase
of N with C for all the surfactants investigated, irrespective of the value of x [15]. The
variations of N with the spacer length, here the total number nT of oxygen and carbon
atoms in the spacer, for the 12-EOx-12 and 16-EOx-16 surfactants are represented in Fig.
4. The difference of behavior between the two series of surfactants is clearly seen. Note
that the values of N are obtained from a fairly complex analysis of SANS data [15,20,21].
Some subtle differences may exist between the methods of analysis used in the two sets
of studies and might be responsible in part for the observed difference in behavior.
Figure 4 and Table 1 show that the values of N for m-EOx-m surfactants are larger than
for m-s-m surfactants. This behavior correlates well with the larger value of the packing
parameter P for the former. This larger value of P arises from the smaller value of the
surface area a of m-EOx-m surfactants [15] relative to m-s-m surfactants [40], for a given
spacer length (i.e., for s=nT =3x+2), which, in turn, mainly reflects a difference between
the conformations of the (CH2)s and EOx spacers in the two series of surfactants. The
Gemini surfactants 150
hydrophilic EOx spacer always remains in contact with water, whereas the (CH2)s spacer
is located at the micelle-water interface for low values of s and in the micelle for large
values of s.
3.
Nonionic Gemini Surfactants
A few investigations concerned the size or aggregation number of nonionic gemini
surfactants [25,26,31,47]. The dimensions of the micelles of the di-chained disugar
surfactants (CmH2m+1)2C[(CH2NHCO(CHOH)4CH2OH]2 have been determined by SANS
and found to depend substantially on m [25,26]. One study clearly showed that the N
values for the nonionic gemini surfactant
{C11H23C(O)N[(CH2CH2O)5CH2CH2OH}2(CH2)2 are much larger than for the
corresponding monomeric surfactant C11H23C(O)N-(CH3)(CH2CH2O)5CH2CH2OH, as for
the ionic gemini (see Fig. 6) [31]. However, this result has largely to do with the fact that
the cloud temperature of the gemini surfactant is lower than that of the corresponding
monomeric surfactant (see Chapter 6, Section II.C). The diameter of the micelles of the
asymmetric gemini surfactant C8H17CH[O(CH2CH2O)16CH3]-CHOHC7H14CN has been
found to be around 4 nm. This value is rather small, indicating a rather low aggregation
number that is probably due to the large size of the head group [47].
4.
Oligomeric Surfactants
The N values for a few oligomeric surfactants have been determined. The aggregation
number of the trimeric surfactant 12–3–12–3–12 was found to be close to that of 12–2–12
and much larger than that of the corresponding
C.
Effect of Additives on the Aggregation Number of Gemini Surfactant
Micelles
The effect of various types of additive on the state of aggregation of gemini surfactants
has been investigated. Often, the effect of an additive was found to be qualitatively
similar to that of the same additive on the aggregation of the corresponding conventional
surfactant.
The effect of cyclodextrin on the aggregation number of the micelles of the
asymmetric nonionic gemini surfactant C8H17CH[O(CH2CH2O)16CH3]-CHOHC7H14CN
has been investigated by self-diffusion nuclear magnetic resonance (NMR) [48] and
SANS [49]. The results suggested the formation of short rodlike micelles and permitted
the determination of their size.
The size of 12-s-12 micelles was shown to be decreased in the presence of
poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) triblock copolymers
[50].
The effect of the addition of the nonionic polymers hydroxypropylguar (HPG) and
hydrophobically modified HPG on the aggregation number of 12-s-12 surfactants has
been investigated [51]. The aggregation number of the polymer-bound surfactant was
well below that in the absence of the polymer and increased with the surfactant
concentration. This behavior is similar to that reported for conventional surfactant-
nonionic polymer systems.
The addition of sodium hyaluronate to solutions of 12-s-12 surfactants, in the presence
of a high concentration of NaCl in order to prevent precipitation of a polymer/12-s-12
complex, was shown to result in a rapid increase of the surfactant aggregation number
Gemini surfactants 152
[52]. Again, this behavior is similar to that noted for conventional surfactant-oppositely
charged surfactant systems.
D.
Polydispersity of Gemini Surfactant Micelles
Information on the polydispersity of micelles of dimeric surfactants has been obtained
from TRFQ [29] and static and dynamic light scattering [32,34,53] studies of m-s-m
surfactants. Recall that the aggregation number measured by TRFQ is independent of the
quencher concentration [Q] for monodisperse micelles but is a decreasing function of this
concentration in the case of polydisperse micelles [54]. The weight-average aggregation
number and the standard deviation are obtained by fitting an appropriate expression of N
to the experimental variation of N with [Q] [54]. SANS is also capable of providing
information on the micelle polydispersity [35]. However, SANS studies of gemini
surfactant solutions all assumed monodisperse micelles [15–21].
Time-resolved fluorescence quenching studies showed a rather small decrease of N
upon increasing quencher concentration for a 15-mM solution of 12–3–12 and 50-mM
solutions of 12-s-12 and 12–6–12, revealing a very small polydispersity of the
quasispherical micelles present in those systems. Such a behavior is in agreement with
the theoretical prediction that spherical micelles are nearly monodisperse [39].
Significant decreases of N upon increasing quencher concentration were observed for the
55-mM 12–3–12, 102-mM 12–4– 12, and 253-mM 12-s-12 solutions. This indicated
micelle polydispersity, in line with the larger values of N in those systems, all above 50
(i.e., 100 dodecyl chains), revealing nonspherical micelles [29]. Thus, as dimeric
surfactant micelles grow, their shape changes and their polydispersity increases.
The polydispersity of m-4-m [32] and of 14–2–12 [34] micelles has been investigated
by dynamic light scattering. In both studies, the authors reported a bimodal distribution of
micelle diameters even at very low concentration, close to the cmc, for all m-4-m
surfactants, even for the short-chain gemini surfactant with m=8. The data were
interpreted as indicating the coexistence of small spherical micelles with large aggregates
of size around 200 nm. Such a result is surprising at least for the 8–4–8 gemini surfactant.
Only the large aggregates remained in solutions of gemini surfactants m-s-m with m>12
[32]. A study of salt-free 12–2–12 solutions was performed that combined x-ray
scattering and static and dynamic light-scattering measurements in the range from the
cmc on to about 2 wt% surfactant [53]. The results indicated the coexistence of small,
intermediate, and large micellar sizes in the whole concentration range. These results
agree with those reported in a study of 12–2–12 solutions by transmission electron
microscopy at cryogenic temperature described in Section IV.B.1.
Properties of micelles and of micellar solutions of gemini (dimeric) surfactants 153
IV.
MICELLE SHAPE AND MICROSTRUCTURE IN SOLUTIONS OF
GEMINI SURFACTANTS
A.
Methods for the Study of Micelle Shape and Microstructure of
Solutions
Essentially two techniques have been used to obtain information on the shape of micelles
of gemini surfactants and on the microstructure in the solutions of these surfactants. The
first technique is the SANS, where the information on the micelle shape is obtained from
the fitting of theoretical expression of the scattered intensity I that depend on the shape of
the scattering particles to the plots representing the variations of I with the scattering
vector [35]. As in other scattering techniques, the quality of the fit does not provide a
certainty concerning the assignment of the micelle shape. The second technique, the cryo-
transmission electron microscopy (cryo-TEM), provides a direct visualization of the
micelles present in thin layers of vitrified specimen of the investigated solutions [55]. The
specimens are prepared by fast-freezing and virification of a thin layer of solution
deposited on an electron microscope grid and quickly immersed in liquid ethane at the
temperature of liquid nitrogen. The resolution of the cryo-TEM technique has
dramatically improved in recent years with the digital recording of the images. This
permits imaging at very low electron exposure to select the appropriate areas on the
specimen and recording the images at much higher magnification than was possible with
photographic films. Artifacts may occur during the preparation of the specimen for cryo-
TEM (blotting of the solution and vitrification). However, these are substantially reduced
compared to conventional electron microscopy because cryo-TEM involves no drying or
staining of the investigated solution and because freezing is performed very quickly,
contrary to the same operation in freeze-fracture electron microscopy.
Light scattering can also be used for studying micelle shape [36], but it is less
powerful than neutron scattering [35].
In this section, a distinction is made between micelle shape and microstructure in the
solution. The latter refers to a higher level of organization, such as entangled threadlike
micelles, multiconnected threadlike micelles, multilamellar vesicles, and so forth.
The microstructure of mixed solutions of gemini and conventional surfactants is
reviewed in Chapter 10, Section IV.
B.
Effect of the Surfactant Chain Length and of the Length and Nature
of the Spacer on the Micelle Shape
The most extensive reports concern the m-s-m and m-EOx-m surfactants. These studies
give a good picture of the effects of the surfactant chain length and of the nature and
length of the spacer on the micelle shape and microstructure in the solution.
Gemini surfactants 154
1.
m-s-m Surfactants
Small-angle neutron scattering and cryo-TEM studies of the m-s-m gemini surfactants all
revealed that the “dimerization” of conventional surfactants always results in the
formation of less curved aggregates, at least when the spacer, whether hydrophilic or
hydrophobic, is short enough. This rule holds even better (i.e., for longer spacers) when
the comparison involves a trimeric or a tetrameric surfactant and the corresponding
conventional surfactant.
The results reported for m-s-m surfactants show that the effect of the alkyl chain
length on the micelle shape is qualitatively similar to that for conventional surfactants.
That is, an increase of alkyl chain carbon number m results in less curved surfactant
aggregates. For instance, the micelles of 10–3–10 are spheroidal [17,18], those of 12–3–
12 are spheroidal at low concentration but elongated at higher concentration [29], and
those of 16–3–16 are disklike [19]. The effect of the nature of the ionic head group
appears to be not too important. Thus, both the cationic 16–3–16 and the anionic 16–2–
16,2Na+ [20] form disklike micelles.
The effect of the spacer length is unexpected. Consider, for instance, the micellar
solutions of surfactants 12-s-12 with increasing spacer carbon number s. Cryo-TEM
showed that the 12–2–12 micelles are threadlike at a concentration of 2 wt%, whereas
those of 12–3–12 are still spheroidal at a concentration of 3 wt% but become threadlike at
a concentration of 7 wt% [29]. 12-s-12 surfactants with 4 < s≤12 all form spheroidal
micelles [29]. At s ≥14, the surfactant forms vesicles, which are often doubly lamellar for
12–20–12 [29]. Note that the quaternary ammonium bromide surfactants with two
unequal alkyl chains, 12–8 and 12–10, which can be considered as the monomers
corresponding to 12–16–12 and 12–20–12, also form vesicles. Thus, upon increasing s
(i.e., as the 12-s-12 surfactant becomes more hydrophobic), the unusual sequence of
micelle shape is observed:
Threadlike micelles → Spheroidal micelles → Vesicles (bilayers)
The 10-s-10 surfactants behave similarly to the 12-s-12 surfactants [18]. The 10–2–10
surfactant forms elongated micelles, whereas the surfactants with 4≤s≤12 give rise to
spheroidal micelles. Surfactants with a longer spacer were not investigated.
The 16-s-16 surfactants show a behavior even more complex than 12-s-12 surfactants.
Cryo-TEM investigations revealed a mixture of vesicles, bilayers fragments, and
threadlike micelles for 16–3–16, threadlike micelles for 16–4–16, still slightly elongated
micelles for 16–6–16, and spheroidal micelles for 16–8–16 [29]. These results were
supported by a SANS study that also showed that 16–12–16 micelles are spheroidal [16].
The surfactant cetyltrimethylammonium bromide (CTAB), which can be considered as
the monomer of 16–3–16, forms elongated micelles only at high concentration. Thus, as
for the 12-s-12 series, “dimerization” of conventional surfactants with m=16 results in
aggregate of lower curvature for short spacers. The sequence of aggregate shape for 16-s-
16 surfactants upon increasing s is
Vesicle+Threadlike micelle+Bilayer fragment → Threadlike micelle → Spheroidal
micelle
The aggregates of 16-s-16 would probably be disklike or vesicular at s > 12, but these
surfactants have not yet been investigated. The lengthening of the
Properties of micelles and of micellar solutions of gemini (dimeric) surfactants 155
polymethylene spacer group for the anionic surfactants 16-s-16,2Na+ results in the same
sequence of micelle shape as for the 16-s-16 surfactants [20]. Thus, the nature of the
electrical charge of the head groups appears to be of little importance in determining the
shape of gemini surfactant micelles, a behavior similar to that of conventional surfactants.
The above differences of behavior between gemini surfactant aggregates and the
corresponding conventional surfactant aggregates can be explained in terms of the
Gemini surfactants 158
packing parameter P[39]. Consider the 12-s-12 surfactants. At the air-water interface, the
surface area a per amphiphilic moiety of 12–2–12 is about 0.40 nm2 [8] as compared to
0.62 nm2 for DTAB [56]. The calculated value of P for 12–2–12 is above 1/3 and
corresponds to elongated micelles, whereas the lower value of P for DTAB corresponds
to spherical micelles, as experimentally observed. The sequence of micelle shape for 12-
s-12 surfactants upon increasing s is also well explained in terms of the value of P and of
location and conformation of the spacer. Indeed, short spacers (s≤8) remain at the micelle
surface [41]. Because a increases rapidly with s at low value of s [40], with v and l nearly
unchanged, P will first decreases upon increasing s, resulting in a change of micelle
shape from elongated to spherical/spheroidal, as experimentally observed. At s>10–12,
the spacer penetrates in the micelle core. The hydrophobic moiety then includes the
spacer and, therefore, its volume v increases with s. The increase of v and the decrease of
a upon increasing s for s>12 [40] result in a rapid increase of P. Thus, the corresponding
aggregates should be less curved than spheroidal micelles. Cryo-TEM reveals that,
indeed, these aggregates are vesicles.
Cryo-TEM permitted a detailed study of the growth of 12–2–12 micelles upon
increasing concentration and of the change of microstructure of the solution associated
with this growth (see Fig. 7) [57]. The micelles in the 0.26 wt% solution are essentially
spherical (Fig. 7A). Threadlike micelles coexist with spheroidal micelles in the 0.5 wt%
solution (Fig. 7B). As the concentration is increased from 0.5 to 1 wt%, the fraction of
surfactant under the form of threadlike micelles increases and the number of spherical
micelles per unit volume decreases rapidly (Figs. 7C and 7D). At 1 wt%, branched
threadlike micelles as well as closed ring micelles are observed (Fig. 7E). A network of
connected cylindrical micelles and closed ring micelles with few isolated spherical and
closed ring micelles are observed at 1.5 wt% (Figs. 7F and 7G). The high resolution of
the method showed that the diameter of the end caps of the threadlike micelles was larger
than the diameter of the cylindrical part of the micelles (inset in Fig. 7D), in agreement
with theoretical predictions [57].
The micelle shape and microstructure in solutions of the dissymmetric gemini
surfactants m-2-m′ has been investigated using various techniques [58,59]. These
surfactants are identical to m-s-m surfactants except for the fact that the two alkyl chains
have different carbon numbers. Surfactants with small values of m+m′ or large values of
m—m′ formed spherical or wormlike micelles. Surfactants with large values of m+m′ or
small values of m—m′ gave rise to lamellar and tubular structures (see Fig. 8) [58].
The shape of the micelles of two trimeric surfactants and one tetrameric surfactant of
structure similar to that of m-s-m surfactants has also been investigated by cryo-TEM.
The solutions of the trimeric surfactant 12–3–12–3– 12 revealed the first branched
threadlike micelles ever observed with a surfactant in the absence of any additive [6,60].
The trimeric surfactant 12–6–12–6–12 formed spheroidal micelles [6]. Thus, the effect of
the spacer length on the micelle shape is qualitatively similar for the dimeric and trimeric
surfactants. The solution of the tetrameric surfactant 12–3–12–4–12–3–12 was found to
contain a significant fraction of surfactant under the form of isolated
Properties of micelles and of micellar solutions of gemini (dimeric) surfactants 159
orientations in the micelles. This possibility decreases the energy cost for branching of a
thread-like micelle.
An intriguing problem that arises with cylindrical micelles of gemini surfactants is that
of the orientation of the spacer with respect to the axis of the micelle (spacer azimut) and
the angle that the spacer makes with the
2.
m-EOx-m Surfactants
Studies of shape of micelles of m-EOx-m surfactants were also performed by SANS and
cryo-TEM. Disklike micelles were not observed with m-EOx-m surfactants. Indeed, even
the 16-EO1-16 surfactant only forms cylindrical micelles [20,21] and so do 16-EO2-16
and 16-EO3-16. The shorter-chain surfactant 12-EO1-12 forms cylindrical micelles [15].
The results reported for this surfactant emphasize the shortcomings of the models used in
the analysis of SANS data for micellar solutions [15]. Indeed, cryo-TEM showed that the
10 wt% solution of 12-EO1-12 contains extremely long threadlike micelles, some of them
apparently showing branching. Nevertheless, the analysis of the SANS data for this
solution was performed assuming monodisperse elongated micelles. The fit to the data
was extremely good and yielded a value of 273 for the aggregation number.
Unfortunately, this value does not provide even a qualitative account for the cryo-TEM
observation.
The effect of the spacer chain length on the shape of m-EOx-m micelles is quite
different from that for m-s-m micelles. Indeed, the 12-EOx-12 micelles are threadlike for
x=1 and 2 and spherical for x=3 and 4 [15]. This change of shape is qualitatively similar
to that observed for 12-s-12 micelles upon increasing s. However, the change of shape for
12-s-12 micelles takes place at s>3 (see Section IV.B.1). In terms of spacer length, s=3
corresponds to x<1. The shape change occurs for a longer spacer for 12-EOx-12 than for
12-s-12 surfactants because the surface area a per amphiphilic moiety is smaller for the
former, at a given value of the spacer length (i.e., at s=3x+2). For instance, the values of a
for the 12–8–12 and 12-EO2-12 surfactants are respectively 0.88 nm2 [40] and 0.6 nm2
[15] per amphiphilic moiety. The micelles of 16-EOx-16 surfactants with x=1–3 are all
elongated [21], whereas the micelles of 16–8–16 and 16–12–16 surfactants are spheroidal
[16,19]. In addition, the 16-EO3-16 micelles are more elongated than the 16-EO2-16
micelles [21]. The differences in values of a between m-s-m and m-EOx-m surfactants
arise from differences in conformation of the spacer. The oxygen atoms in the spacer of
m-EOx-m surfactants tend to maintain the spacer in the aqueous phase. This, in turn,
favors more compact conformations of the surfactant molecule and thus lower values of a
and aggregates of lower curvature than for m-s-m surfactants. This reasoning is supported
by the fact that the average distance between two head groups is smaller in 16-EOx-16
micelles than in 16-s-16 micelles [21].
3.
Nonionic Gemini Surfactants
The nonionic gemini surfactants show the same sequence of micelle shapes as ionic
gemini surfactants. However, owing to the much weaker repulsive electrostatic
interactions between nonionic gemini head groups, both elongated and disklike micelles
or vesicles are found to occur at shorter alkyl chain lengths than for ionic gemini
surfactants.
The most comprehensive study of nonionic gemini surfactants concerns the
diglucamide surfactants (CmH2m+1)2C[CH2NHCO(CHOH)4CH2OH]2 [25,26]. A SANS
study showed that the micelles are slightly elongated for the surfactant with m=5 (aspect
ratio=2), rodlike for the surfactants with m =6 and 7, and disklike for the surfactant with
Gemini surfactants 162
m=8 [26]. Recall that the formation of rodlike and disklike micelles requires short spacers
and values of m of 10–12 and 16, respectively, for m-s-m surfactants.
The surfactants [C13H27C(O)NCH2(CHOH)4CH2OH]2(CH2)s were investigated by
cryo-TEM in the presence of 5 mol% anionic surfactant, added to increase their solubility
in water [64]. The surfactants with s=6 and 8 formed threadlike micelles coexisting with
a few vesicles and bilayer fragments, whereas the surfactant with s=10 formed vesicles.
This behavior is different from that of the ionic 12-s-12 gemini surfactant micelles, which
are all spheroidal for s=6–10 and show a change of shape from spheroidal to vesicular
upon increasing s from 10 to 16.
The asymmetric surfactant C8H17CH(OH)CH[O(CH2CH2O)16CH3] C7H14CN also
forms nearly spherical and slightly polydisperse micelles [49]. The micelles of
Tyloxapol, a polydisperse mixture of oligomers of Triton X-100, are spherical [28].
4.
Other Gemini Surfactants
The formation of vesicles and of bilayer fragments has been reported for many different
gemini surfactants in water. The surfactants [C16H33N+(CH3)2, C15H31CO−2](CH2)s, with
s=2–12 and where the counterion is a stearate anion, form vesicles [65].
Dimeric surfactants with dimethylammonium bromide head groups, a polymethylene
spacer group with s=32 and alkyl chains with m=16, form vesicles in which a significant
fraction of the surfactant adopts a conformation with one head group on the vesicle outer
layer and the other on the vesicle inner layer [66,67]. These compounds have been
referred to as bilayer-bridging lipids [67]. Similar gemini surfactants with phosphate
head groups give rise to lamellar aggregates and vesicles [68–70]. Their study also led to
the conclusion that part of the aggregated surfactant acts as bilayer-bridging lipids. All of
these compounds constitute good models for the lipids found in the membrane of
Thermoacidophilus archaebacteria that can survive at extreme temperatures and pH.
This stability is due to the fact that each alkyl chain connects the two head groups and
thus spans the whole membrane.
True dimeric lipids, where each hydrophobic moiety includes two alkyl chains, such
as [(C16H33OCH2)CH(OC16H33)CH2N+(CH3)2,Br−]2(CH2)s, were shown to give rise to
different types of bilayer organization [71–73].
Studies of several other gemini surfactants report formation of vesicles: diphosphate
surfactants with s=24 and m=12 (the long spacer group may be bilayerspanning) [74];
gemini surfactant with a cyclic spacer, two potassium carboxylate head groups and
pentadecyl chains [75]; surfactants with partially hydrogenated and partially fluorinated
alkyl chains [76,77]; asymmetric gemini surfactants
CmH2m+1N+(CH3)2CH2CONHN=C(CH2CH2 CO2−) CmH2m+1 [78]; surfactants with two
head groups and three alkyl chains [79].
A peptide-based dimeric surfactant with m=11 was reported to give rise to fibrils and
ribbons with a periodic twistlike structure [80]. These structures probably arise owing to
the presence of asymmetric carbons in this surfactant. Vesicles and helical structures
were also reported in mixtures of a lipid derived from histidine and a phosphatidic acid-
based gemini surfactant that contained two asymmetric carbon atoms [81]. Helical fibers
were also observed in organic solvents such as toluene, xylene and pyridine with the
Properties of micelles and of micellar solutions of gemini (dimeric) surfactants 163
5.
Change of Shape of Gemini Surfactant Aggregates
Several studies showed that changes of shape of gemini surfactant aggregates can be
induced in different ways, depending on the nature of the surfactant. Thus, the disklike
micelles of 16–3–16 are transformed into threadlike micelles when increasing the
temperature from 30 °C to 45 °C [89]. The nonionic sugar-amine surfactant hexane-1,6-
bis(hexadecyl-1-deoxyglucityl-amine) forms micelles at pH<4.0 and vesicles in the pH
range 5.5–7.5 [90]. Shearing of solutions of 12–2–12 threadlike micelles induced
anisotropy at above a certain shear rate and resulted in the formation of aggregates of
micelles [91].
Additives can also induce changes of aggregate shape. Addition of the cosurfactant n-
hexanol to a micellar solution of 12–2–12 threadlike micelles results in the formation of
vesicles [92]. Additions of spherical micelle-forming surfactants such as DTAB and 12–
10–12 to the doubly lamellar vesicles of 12–20–12 bring about the progressive
transformation of the vesicles into spheroidal micelles [93]. Additions of the threadlike
micelle-forming 12–2– 12 to 12–20–12 vesicles results in the progressive breakage of the
vesicles and their transformation into branched threadlike micelles [94]. The
transformation in the intermediate range of composition is extremely slow and stretches
over weeks.
V.
MICROPOLARITY AND MICROVISCOSITY OF GEMINI
SURFACTANT MICELLES
A.
Micropolarity
The micropolarity of gemini surfactant micelles has been much investigated using pyrene
as the fluorescent probe. The ratio I1/I3 of the intensities of the first and third peaks in the
pyrene emission spectrum is usually taken as a measure of the polarity of the pyrene
microenvironment [3–6,21,28,31,47, 50,51,99–101]. Indeed, this ratio takes the values
1.85 in water, about 0.6 in alkanes, and intermediate values in solvents of intermediate
polarity [102]. Note that pyrene is preferentially located in the palisade layer of micelles
[37]. Two studies of gemini surfactants used pyrenecarboxaldehyde as the fluorescent
probe [103,104]. The wavelength at maximum intensity of the fluorescence emission
spectrum of this probe is sensitive to the polarity of the probe environment [37].
The variation of the I1/I3 ratio with the surfactant concentration usually shows a
sigmoidal decrease from a value equal to that in water at C<<cmc, to a nearly
concentration-independent value that characterizes the micelle polarity at C>>cmc
[37,102]. This property has been much used for the determination of the cmc of gemini
surfactants [3,5,28,31,47,50,51,99,100] and also for obtaining the value of I1/I3
characterizing the micropolarity of gemini and oligomeric surfactant micelles
[4,6,21,101].
The variation of the I1/I3 ratio with the spacer carbon number for the 12-s-12
surfactant series has been investigated and compared to that for the corresponding
monomeric surfactants dodecyl(alkyl)dimethylammonium bromide (denoted 12-s/2). The
results are shown in Fig. 11. The I1/I3 ratio for the gemini surfactants is a maximum at s
around 5–6, as for the cmc and some of the thermodynamic properties reviewed in
Gemini surfactants 166
Chapter 6, Sections III and IV, and probably for the same reason. For s>4, a decrease of
I1/I3 (i.e., of micropolarity) is observed for the 12-s-12 and 12-s/2 surfactants, upon
increasing s. In addition, the results for the two series of surfactants nearly fall on the
same curve. Thus, pyrene apparently reports the same micropolarity in a gemini
surfactant micelle and in a micelle of the corresponding monomeric surfactant. This
result has been attributed to the fact that pyrene is located in the palisade layer of micelles
of cationic surfactants and that the palisade layer of 12-s-12 and 12-s/2 surfactant have
nearly the same composition [6,101]. The decrease of polarity upon increasing value of s
is due to the progressive replacement of some water molecules from the palisade layer by
methylene groups of the spacer (12-s-12 surfactants) or of the variable alkyl chain (12-s/2
surfactants) as s is increased.
micelles of the monomeric and oligomeric surfactants do not differ much and, thus,
similar values of the micropolarity are reported by pyrene. Similar values of I1/I3 at
C>>cmc have been reported for the gemini surfactant [C11H23C(O)NCH2CH2(OCH2-
CH2)5OH]2(CH2CH2) and the corresponding monomeric surfactant
C11H23C(O)N(CH3)CH2CH2(OCH2CH2)5OH [31]. The values of I1/I3 for the micelles of
the nonionic surfactants Triton X-100 and Tyloxapol, a mixture of oligomer of Triton X-
100, are also quite close at 25°C and 40°C [28].
All the above studies referred to gemini or oligomeric surfactants with a hydrophobic
spacer. A study of the gemini surfactants 16-EOx-16 with a hydrophilic spacer showed no
effect of the number of ethylene oxide units on the value of the I1/I3 ratio [21]. This result
confirms the different behaviors of m-EOx-m and m-s-m surfactants. The hydrophilic
spacer of the m-EOx-m surfactants does not penetrate in the micelles and the
micropolarity of the palisade layer is therefore very little affected by the spacer length.
On the contrary, when it is long enough, the hydrophobic spacer of m-s-m surfactants
penetrates into the micelles and thereby decreases the polarity sensed by pyrene.
The addition of water-soluble polymers to gemini surfactant solutions left the
micropolarity unchanged (m-s-m/hydrophobically modified guar systems) [51] or
decreased it (12-s-12/Pluronic® triblock copolymers) [50].
B.
Microviscosity
The microviscosity of gemini surfactant micelles was investigated using the fluorescent
probes dipyrenylpropane (measurement of the ratio IM/IE of the intensities of the
monomer and excimer emissions) [101,105] and diphenylhexatriene (measurement of the
fluorescence anisotropy) [16,21,101].
The microviscosity and the activation energy for the microviscosity increase with the
gemini surfactant chain length as for conventional surfactants [101,105].
The microviscosity of 12-s-12 and 16-s-16 surfactant micelles decreases rapidly at
s>4, as shown in Fig. 12. On the contrary, the microviscosity of the monomeric
surfactants 12-s/2 shows almost no dependence on s and is always lower than for the
corresponding dimeric surfactants 12-s-12 (see Fig. 12). The effect of the spacer carbon
number has been discussed in terms of micelle shape. The 12–2–12 and 12–3–12
surfactants tend to form elongated micelles at the concentration used in the study,
corresponding to the results in Fig. 12. The tighter packing of the surfactants in such
micelles is responsible for the higher values of the microviscosity [101].
The microviscosity of the 16-EOx-16 micelles varies little with the spacer length [21].
This result reflects, once again, the importance of the spacer conformation on the micellar
properties.
The microviscosity increases with the surfactant concentration C for the gemini
surfactants that undergo a sphere-to-rod transition [101]. Thus, the microviscosity of 12–
6–12 is independent of C, whereas that of 12–2–12 and 12–3–12 increases with C and
then levels out. At very low concentration, close to the cmc, the micelles of these three
surfactants have nearly the same microviscosity. This behavior is in agreement with the
changes of micelle aggregation number of the three surfactants (see Section III.B.1 and
Fig. 3).
Gemini surfactants 168
The microviscosity of oligomeric surfactant micelles has been found to increase nearly
linearly with the degree of oligomerization z (number of alkyl chains) for two series of
quaternary ammonium oligomeric surfactants with dodecyl chains (m=12, see Fig. 13)
[6,101]. This result has been explained by the fact that the motion of a fluorescent probe
in a micelle involves the displacement of surfactant alkyl chains. This motion becomes
increasingly
VI.
DYNAMICS OF GEMINI SURFACTANT MICELLES
Micelles are dynamic objects that are constantly exchanging surfactant with the
intermicellar solution (surfactant-exchange process) and that are forming and breaking
up, as schematically represented in Fig. 14 [106,107]. Several processes can contribute to
the micelle formation/breakup [108]. This also applies to gemini surfactants.
Gemini surfactants 170
The dynamics of gemini surfactant micelles has been mainly investigated by means of
chemical relaxation techniques [109]—more specifically, ultra-
of TR are of several milliseconds for a gemini surfactant with m=12. Such values are
comparable and even larger than for a conventional surfactant with m=16 [115]. In
addition, the value of k+ for the 12-s-12 surfactants increases with the spacer carbon
number (see Table 2). This result is rather unexpected because k+ usually remains
constant or decreases slightly when the surfactant becomes more hydrophobic [115]. This
behavior of k+ may be related to the conformational change of the surfactant that occurs
as s is increased from 2 to 4 (see Chapter 5, Section V). Additional results that extend the
range of
TABLE 2 Rate Constants for the Entry in and Exit
from Micelles of m-s-m Surfactants
Surfactant 10−7k+(M−ls−1) 10−4k−(s−1) Ref.
8–3–8 160 8,800 3
8–6–8 230 16,000 3
12–2–12 3.1 2.8 110
12–3–12 4.3 4.0 110
12–4–12 8.7 9.5 110
Note: Values of the rate constants at 25°C for 8-s-8 and 15°C for 12-s-12 surfactants.
variation of k+ with s are required in order to have a better insight in the origin of the
variation of k+.
Nuclear magnetic resonance studies of dimeric surfactants with still longer alkyl
chains (14–2–14, assymmetric surfactants m-2-m′ with m+m′ >28 and C8F17C4H8-2-
C4H8C8F17 surfactants) [77,112] confirmed the above trends. The experiments showed
separate signals for the free and aggregated surfactants, revealing that the exchange
between free and aggregated states was slow on the NMR timescale. It must be pointed
out that, in these experiments, the surfactant was under the form of elongated micelles
and/or vesicles. This may have resulted in a significant slowing down of the surfactant
exchange with respect to that with spherical micelles.
The dynamics of the micelle formation/breakup in solutions of 12-s-12 surfactants was
investigated by the pressure-jump technique [110]. The measurements were restricted to
concentrations ranging from slightly above the cmc to about 3 cmc, owing to
experimental constraints. This range corresponds to spherical micelles (see Section
III.B.1). The variation of the relaxation time τ2 associated to the micelle
formation/breakup revealed that this process occurs via stepwise entry/exit reactions of
one surfactant at a time in/from micelles. The values of τ2 were in some instances larger
than 1 s. A temperature-jump study of the surfactant 12–2–12 in the presence of NaF also
yielded large values of τ2 [111]. The values of the micelle lifetime TM for 12-s-12
surfactants were obtained from the values of τ2 using the expression of TM derived by
Aniansson [116]. The values of TM were found to be quite large, reaching tens to
hundreds of seconds.
From the above, it is clear that the surfactant residence times in, and the lifetimes of,
micelles are much longer for gemini than for conventional surfactants. This fact has the
Gemini surfactants 172
VII.
SOLUBILIZATION BY MICELLAR SOLUTIONS OF GEMINI
SURFACTANTS
A.
m-s-m Surfactants
A comprehensive study of the solubilization of a typical aromatic molecule, the
transazobenzene, in micellar solutions of 10-s-10 and m-6-m surfactants was reported by
Devinsky et al. [120]. The values of the solubilization capacity SC of the surfactants were
Properties of micelles and of micellar solutions of gemini (dimeric) surfactants 173
calculated from the reported values of the solubility and of the cmc. The variations of the
solubilization capacity SC of the surfactant with the spacer carbon number s for 10-s-10
surfactants and with m for the m-6-m surfactants are represented in Fig. 15. It is seen that
SC varies nearly linearly with m that is, with the volume of micellized surfactant, a
behavior similar to that for conventional surfactants. On the other hand, SC is a maximum
at around s=6, that is, a value of s close to that at which the cmc
and the increasing hydrophobicity of the spacer. The maximum of solubilization capacity
also correlates with the minimum of packing parameter P discussed in Section IV.B.1.
However, the s dependence of the solubilization capacity of micellar solutions of m-s-
m surfactants for aromatic molecules seems to depend heavily on the system investigated.
The solubilization capacity of 12-s-12 surfactant micelles for the aromatic molecule
styrene is represented in Fig. 16. At a given surfactant concentration, between 5 and 10
wt% (i.e., 80–150 mM), the solubilization capacity increases rapidly, with s between 2
and 3, then more slowly up to s=12 [121]. No explanation can be given presently for this
difference in behavior noted between the trans-azobenzene/10-s-10 and the styrene/12-s-
12 systems. Differences in micelle shape cannot be invoked
Dam et al. [122] also showed that micellar solutions of m-s-m surfactants have better
solubilizing properties than comparable solutions of conventional surfactants (see Table
3). In addition, they showed that m-s-m surfactants as well as the cationic conventional
surfactant CTAB have a larger solubilizing capacity for toluene, an aromatic molecule,
than for n-hexane (see Table 3). For both CTAB and the m-s-m surfactants, this behavior
reflects the attractive interaction that exists between quaternary ammonium head groups
and the π-electron cloud surrounding aromatic molecules [127–129]. Note that the results
listed in Table 3 for 12-s-12 surfactants show a decrease of the solubilization capacity as
s is increased from 2 to 10, a variation opposite to that seen in Fig. 16. Unfortunately, the
concentration of the surfactant in those experiments was not specified.
TABLE 3 Solubilization Capacity (in mol/mol) of
Micellar Solutions of Gemini Surfactants for
Toluene and n-Hexane
Surfactant SC (toluene) SC (n-hexane)
10–2–10 3.21 0.62
12–2–12 3.50 1.03
12–6–12 2.40 —
12–10–12 2.13 —
CTAB 0.59 0.21
SDS (sodium dodecyl sulfate) 0.25 0.35
Source: Ref. 122
B.
12-EOx-12 Surfactants
The solubility of styrene in relatively concentrated solutions of 12-EOx-12 surfactants has
been determined in view of polymerizing styrene in these systems [15]. The results at
60°C represented in Fig. 17 can be directly compared to those for the solubilization of
styrene in micellar solutions of 12-s-12 (Fig. 16). In both instances, for spacers long
enough (s>4, x>1), the solubilization is larger in gemini surfactant solutions than in
solution of the corresponding monomeric surfactant DTAB. Also, in both instances, the
solubilization capacity increases very rapidly between s=2 and s=3 for 12-s-12
surfactants and between x=1 and x=2 for 12-EOx-12 surfactants. However, important
differences also exist between the two sets of results. For instance, at a given surfactant
concentration, say between 5 and 10 wt%, the solubilization capacity of 12-s-12 micelles
increases monotonously with s, whereas that of 12-EOx-12 surfactants goes through a
maximum at x between 1 and 2. Also, the plots of the solubilization capacity versus
surfactant concentration for the 12–6–12 and 12–8–12 surfactants show a maximum. The
same plots for 12-EOx-12 surfactants with x=1–5 only show a small increase with C. The
authors attributed the different behaviors of 12-s-12 and 12-EOx-12 surfactants to the
Gemini surfactants 176
change in the nature of the spacer [15]. The solubilization power of 12-EOx-12 micelles is
slightly larger at 25°C than at 60°C [15,123].
C.
Gemini Sugar Surfactants
The solubilization of toluene and n-hexane in micellar solutions of the gemini sugar
surfactants [Cm−1H2m−1CONCH2(CHOH)4CH2OH]2(CH2)s, referred to as mS-s-mS, has
been investigated [124,125]. The effect of the length of the alkyl chain and of the spacer
on the solubilization capacity was investigated using the surfactants series mS-10-mS and
10S-s-10S, respectively, at C=5
then drops to a fairly low value for the 10S-10-10S surfactant, which is the first surfactant
in the 10S-s-10S series that forms vesicles. The effect of the alkyl chain length (Fig. 18,
top) is more complex, with values of SC for toluene larger than for hexane up to m =8,
and then becoming smaller starting at m=12. The authors discussed these variations in
terms of changes of aggregate morphology with m.
VIII.
CONCLUSIONS
This chapter reviewed the main properties of gemini surfactant micelles, namely
ionization, size, polydispersity, shape, dynamics, and solubilization capacity. In a number
of aspects (micelle size, microviscosity, and dynamics; microstructure of the solution)
gemini surfactants showed a behavior remarkably different from that of the
corresponding monomeric surfactants. The results clearly showed the dramatic effect of
the spacer length and nature on the properties of micellar solutions of gemini surfactants
and their microstructure. The length and nature of the surfactant alkyl chain and the
nature of the head group also affect the solution properties but to a much lesser extent
than the spacer group. In the case of oligomeric surfactants, the degree of oligomerization
plays a role nearly as important as the spacer length on the solution properties and
microstructure. However, at the level of chemical synthesis, it is much more difficult and
thus more costly to act on the degree of oligomerization than on the spacer. In addition,
the degree of oligomerization does not offer the modulation of solution properties that
can be achieved by changing the nature of the spacer. Therefore, it is likely that future
research on gemini surfactants will focus on the synthesis of surfactants with functional
spacers: cleavable, photosensitive, pH-sensitive, and so forth. There is in this respect a
nearly infinite number of possibilities just limited by the chemists ability to perform the
required synthesis. Some of these possibilities have started to be exploited. There is no
doubt that they will be further used in the not so distant future.
Gemini surfactants 178
GLOSSARY
REFERENCES
19. Aswal, V.K.; De, S.; Goyal, P.S.; Bhattacharya, S.; Heenan, R.K. Phys. Rev. E. 1998, 57, 776.
20. Aswal, V.K.; De, S.; Goyal, S.; Bhattacharya, S.; Heenan, R.K. Phys. Rev. E. 1999, 59, 3116.
21. De, S.; Aswal, V.K.; Goyal, P.S.; Bhattacharya, S. J. Phys. Chem. B. 1998,102, 6152.
22. Alami, E.; Abrahmsen-Alami, S.; Eastoe, J. and Heenan, R.K. Langmuir 2002, 19, 18.
23. Gaillon, L.; Lelievre, J.; Gaboriaud, R. J. Colloid Interface Sci. 1999, 213, 287.
24. Bales, B. J. Phys. Chem. B 2001, 105, 6798.
25. Eastoe, J.; Rogueda, P.; Harrison, B.J.; Howe, A.M.; Pitt, A.R. Langmuir 1994, 10, 4429.
26. Eastoe, J.; Rogueda, P.; Howe, A.M.; Pratt, A.R.; Heenan, R.K. Langmuir 1996, 12, 2701.
27. Hattori, N.; Hirata, H.; Okabayashi, H.; O’Connor, C.J. Colloid Polym. Sci. 1999, 277, 361.
28. Regev, O.; Zana, R. J. Colloid Interface Sci. 1999, 210, 8.
29. Danino, D.; Talmon, Y.; Zana, R. Langmuir 1995, 11, 1448.
30. Alargova, R.G.; Kochijashky, I.I.; Sierra, M.L.; Zana, R. Langmuir 1998, 14, 5412.
31. Paddon-Jones, G.; Regismond, S.; Kwetkat, K.; Zana, R. J. Colloid Interface Sci. 2001, 243,
496.
32. Pisarcik, M.; Dubnickova, M.; Devinsky, F.; Lacko, I.; Skvarla, J. Colloids Surfaces A. 1998,
743, 69.
33. Pisarcik, M.; Devinsky, F.; Lacko, I. Colloids Surfaces A. 2001, 172, 139.
34. Sikiric, M.; Primozic, I.; Filipovic-Vincenkovic, N. J. Colloid Interface Sci. 2002, 250, 221.
35. Cabane, B. In: Surfactant Solutions. New Methods of Investigation; Zana, R. ed.; New York:
Marcel Dekker, 1987; Chap. 2.
36. Candau, S.J. In: Surfactant Solutions. New Methods of Investigation; Zana, R. ed.; New York:
Marcel Dekker; 1987; Chap. 3.
37. Zana, R.In: Surfactant Solutions. New Methods of Investigation; Zana, R., ed.; New York:
Marcel Dekker; 1987; Chap. 5.
38. Menger, F.M.; Keiper, J.S. Angew. Chem. Int. Ed. 2000, 39, 1906.
39. Israelachvili, J.N.; Mitchell, D.J.; Ninham, B.W. J. Chem. Soc. Faraday Trans. 1976, 72, 1525.
40. Alami, E.; Beinert, G.; Marie, P.; Zana, R. Langmuir 1993, 9, 1465.
41. Alami, E.; Levy, H.; Zana, R.; Skoulios, A. Langmuir 1993, 9, 940.
42. Moroi, Y.; Matuura, R.; Tanaka, M.; Murata, Y.; Aikawa, Y.; Furutani, F.; Kuwamara, T.;
Takahashi, H.; Inokuma, S. J. Phys. Chem. 1990, 94, 842.
43. Missel, P.J.; Mazer, N.A.; Benedek, G.B.; Young, C.Y.; Carey, M.C. J. Phys. Chem. 1980, 84,
1044.
44. Kern, F.; Lequeux, F.; Zana, R.; Candau, S.J. Langmuir 1994,10, 1714.
45. Safran; Pincus, P.; Cates, M.E.; Mackintosh, F. J. Phys. (Paris) 1990, 51, 503.
46. Mackintosh, F.; Safran, S.; Pincus, P. Europhys. Lett. 1990,12, 697.
47. Alami, E.; Holmberg, K. J. Colloid Interface Sci. 2001, 239, 230.
48. Abrahmsen-Alami, S.; Alami, E.; Eastoe, J.; Cosgrove, T. J. Colloid Interface Sci. 2002, 246,
191.
49. Alami, E.; Eastoe, J. J. Colloid Interface Sci. 2002, 255, 403.
50. Wettig, S.D.; Verrall, R.E. J. Colloid Interface Sci. 2001, 244, 377.
51. Kästner, U.; Zana, R. J. Colloid Interface Sci. 1999, 275, 468.
52. Pisarcik, M.; Imae, T.; Devinsky, F.; Lacko, I.; Bakos, D. J. Colloid Interface Sci. 2000, 225,
207.
53. Weber, V.; Narayanan, T.; Mendes, E.; Schosseler, F. Langmuir 2003, 19, 992.
54. Almgren, M. Adv. Colloid Interface Sci. 1192, 41, 9.
55. Talmon, Y. In: Modern Characterization Methods of Surfactant Systems’, Binks, B. ed.; New
York: Marcel Dekker; 1999; Chap. 5.
56. Aveyard, R.; Cooper, P.; Fletcher, P.D. J. Chem. Soc. Faraday Trans. 1990, 86, 211.
57. Bernheim-Groswasser, A.; Zana, R.; Talmon, Y. J. Phys. Chem. B. 2000,104, 4005 and
references therein.
58. Oda, R.; Huc, Y.; Candau, S. J. Chem. Commun. 1997, 2105.
Properties of micelles and of micellar solutions of gemini (dimeric) surfactants 181
59. Oda, R.; Huc, I.; Homo, J.C.; Heinrich, B.; Schmutz, M.; Candau, S.J. Langmuir 1999, 15,
2384.
60. Danino, D.; Talmon, Y.; Levy, H.; Beinert, G.; Zana, R. Science 1995, 269, 1420.
61. In, M.; Aguerre-Chariol, O.; Zana, R. J. Phys. Chem. B. 1999, 103, 7747.
62. Karaborni, S.; Esselink, K.; Hilbers, P.A.; Smit, B.; Karthauser, J.; van Oss, N.M.; Zana, R.
Science 1994, 266, 254.
63. Oda, R.; Laguerre, M.; Huc, Y.; Desbat, B. Langmuir 2002,18, 9659.
64. Pestman, J.; Terpstra, K.R.; Stuart, M.C.; van Doren, H.A.; Brisson, A.; Kellogg, R.M.;
Engberts, J.B.F.N. Langmuir 1997, 13, 6857.
65. Bhattacharya, S.; De, S. Langmuir 1999, 15, 3400.
66. Moss, R.A.; Fujita, T.; Okumura, Y. Langmuir 1991, 7, 2415.
67. Moss, R.A.; Li, J.-M. J. Am. Chem. Soc. 1992, 114, 9227.
68. Kim, J.-M.; Thompson, D.H. Langmuir 1992, 8, 637.
69. Thompson, D.H.; Wong, K.F.; Humphry-Baker, R.; Wheeler, J.J.; Kim, J.-M.; Rananavare, S.
J. Am. Chem. Soc. 1992,114, 9035.
70. Yamauchi, K.; Togawa, K.; Kinoshita, M. J. Biochem. 1996, 119, 115.
71. Bhattacharya, S.; De, S.; George, S. Chem. Commun. 1997, 2287.
72. Dewa, T.; Mikaye, Y.; Kezdy, F.; Regen, S.L. Langmuir 2000,16, 3735.
73. Sugahara, M.; Regen, S.L. Langmuir 2001, 17, 4413.
74. Duivenvoorde, F.L.; Feiters, M.C.; van der Gaast, S.J.; Engberts, J.B.F.N. Langmuir 1997, 13,
3737.
75. Yeager, D.A.; Brown, E.T. Langmuir 1996, 12, 1976.
76. Gaysinski, M.; Joncheray, L.; Guittard, F.; Cambon, A.; Chang, P. J. Fluor. Chem. 1995, 74,
131.
77. Oda, R.; Huc, I.; Danino, D.; Talmon, Y. Langmuir 2000,16, 9759.
78. Yeager, D.A.; Li, B.; Clark, T. Jr. Langmuir 1996,12, 4314.
79. Sumida, Y.; Masuyama, A.; Takasu, M.; Kida, T.; Nakatsuji, Y.; Ikeda, I.; Nojima, M.
Langmuir 2001, 17, 609 and references therein.
80. Jennings, K.; Marshall, I.; Birrell, H.; Edwards, A.; Haskins, N.; Söderman, O.; Kirby, A.J.;
Camilleri, P. Chem. Commun. 1998, 1951.
81. Sommerdijk, N.A.; Lambermon, M.H.; Feiters, M.C.; Nolte, R.J.; Zwanenburg, B. Chem.
Commun. 1997, 1423.
82. Oda, R.; Huc, I.; Candau, S.J. Angew. Chem. Int. Ed. 1998, 37, 2689.
83. Oda, R.; Huc, I.; Schmutz, M.; Candau, S.J.; MacKintosh, F.C. Nature 1999, 399, 566.
84. Berthier, D.; Buffetau, T.; Léger, J.-M.; Oda, R.; Huc, I. J. Am. Chem. Soc. 2002, 124, 13486.
85. Jeager, D.A.; Wang, Y.; Pennington, R.L. Langmuir 2002, 18, 9259.
86. Eastoe, J.; Dominguez, M.S.; Wyatt, P.; Beeby, A.; Heenan, R. Langmuir 2002, 18, 7837.
87. Peresypskin, A.V.; Menger, F.M. Org. Lett. 1999,1, 1347.
88. Menger, F.M.; Peresypskin, A.V.; Caran, K.L.; Apkarian, R.P. Langmuir 2000, 16, 9113.
89. Aswal, V.K.; De, S.; Goyal, P.S.; Bhattacharya, S.; Heenan, R.K. J. Chem. Soc., Faraday
Trans. 1998, 94, 2965.
90. Bergsma, M.; Fielden, M.L.; Engberts, J.B.F.N. J. Colloid Interface Sci. 2001, 243, 491.
91. Oda, R.; Panizza, P.; Schmutz, M.; Lequeux, F. Langmuir 1997,13, 6407.
92. Oda, R.; Bourdieu, L.; Schmutz, M. J. Phys. Chem. B 1997, 101, 5913.
93. Danino, D.; Talmon, Y.; Zana, R. J. Colloid Interface Sci. 1997, 185, 84.
94. Bernheim-Groswasser, A.; Zana, R.; Talmon, Y. J. Phys. Chem. B 2000, 104, 12192.
95. Taupin, C.; Dvolaitzky, M. In: Surfactant Solutions. New Methods of Investigation; Zana, R.
ed.; New York: Marcel Dekker; 1987; Chap. 7.
96. Henderson, C.; Selinger, B.; Watkins, A. J. Photochem. 1981, 16, 215.
97. Hare, F.; Amiel, J.; Lussan, C. Biochem. Biophys. Acta 1979, 555, 388 and references therein.
98. Maiti, N.; Krishna, M.; Periasamy, N. J. Phys. Chem. B 1997, 101, 11051.
99. Esumi, K.; Taguma, K.; Koide, Y. Langmuir 1996, 12, 4039.
Gemini surfactants 182
100. Pinazo, A.; Wen, X.; Pérez, L.; Infante, M.R.; Franses, E.I. Langmuir 1999, 15, 3134.
101. Zana, R.; In, M.; Levy, H.; Duportail, G. Langmuir 1997, 73, 5552.
102. Kalyanasundaram, K.; Thomas, J.K. J. Am. Chem. Soc. 1977, 99, 2039.
103. Esumi, K.; Saika, R.; Miyazaki, M.; Torigoe, K.; Koide, H. Colloids Surfaces A. 2000, 766,
115.
104. Rosen, M.J.; Mathias, J.H.; Davenport, L. Langmuir 1999, 15, 7340.
105. Zana, R. J. Phys. Chem. B. 1999, 103, 9117.
106. Aniansson, E.A.G.; Wall, S.N. J. Phys. Chem. 1974, 78, 1024; 1975, 79, 857.
107. Zana, R. In: Encyclopedia of Surface and Colloid Science; Hubbard, A. ed.; New York:
Marcel Dekker, 2002; p. 1515.
108. Zana, R. In: Dynamic Properties of Interfaces and Association Structures; Pilai, V., Shah,
D.O., eds; AOCS Press: Champaign, IL, 1996; p. 152.
109. Lang, J.; Zana, R. In: Surfactant Solutions. New Methods of Investigation’, Zana, R. Ed.; New
York: Marcel Dekker; 1987; Chap. 8.
110. Ulbricht, W.; Zana, R. Colloids Surfaces A. 2001, 487, 183–185.
111. Oelschlaeger, C.; Waton, G.; Candau, S.J.; Cates, M.E. Langmuir 2002, 18, 7265.
112. Huc, I.; Oda, R. Chem. Commun. 1999, 2025.
113. Candau, S.J.; Hebraud, P.; Schmitt, V.; Lequeux, F.; Kern, F.; Zana, R. Nuov. Cimento 1994,
16D, 1401.
114. Fischer, P.; Rehage, H.; Grüning, B. Tenside Surfact. Detergents 1994, 31, 99.
115. Aniasson, E.A.G.; Wall, S.N.; Almgren, M.; Hoffmann, H.; Kielmann, I.; Ulbricht, W.; Zana,
R.; Lang, J.; Tondre, C. J. Phys. Chem. 1976, 80, 905.
116. Aniansson, E.A.G. Prog. Colloid Polym. Sci. 1985, 70, 2.
117. Cates, M.E.; Candau, S.J. J. Phys. Cond. Mater. 1990, 2, 6869 and references therein.
118. Rehage, H.; Hoffmann, H. Mol Phys. 1991, 74, 933 and references therein.
119. Pinazo, A.; Infante, M.R.; Chang, C.H.; Franses, E.I. Colloids Surfaces A. 1994, 87, 117.
120. Devinsky, F.; Lacko, I.; Imam, T. J. Colloid Interface Sci. 1991, 143, 336.
121. Dreja, M.; Tieke, B. Langmuir 1998,14, 800.
122. Dam, Th.; Engberts, J.B.F.N.; Karthauser, J.; Karaborni, S.; van Os, N.M. Colloids Surfaces
A. 1996, 118, 41.
123. Dreja, M.; Tieke, B. Ber. Bunsenges. Phys. Chem. 1998, 102, 1705.
124. van Doren, H.A.; Smith, E.; Petsman, J.M.; Engberts, J.B.F.N.; Kellogg, R.M. Chem. Soc.
Rev. 2000, 29, 183.
125. Petsmann, J.M. PhD thesis, University of Groningen, 1998.
126. Kim, J.-H.; Domach, M.M.; Tilton, R.D. Langmuir 2000,16, 10037.
127. Bunton, C.A.; Sepulveda, L. J. Phys. Chem. 1979, 83, 680.
128. Lianos, P.; Viriot, M.L.; Zana, R. J. Phys. Chem. 1984, 55, 1098.
129. Kim, K.; Lee, J.Y.; Lee, S.J.; Ha, T.K.; Kim, D.H. J. Am. Chem. Soc. 1994, 116, 7399.
8
Rheology of Solutions of Gemini
Surfactants
MARTIN IN CNRS-Université Montpellier 2, Montpellier, France
I.
INTRODUCTION
Most consumer products used in everyday life are formulations that contain surfactants.
Surfactants are included in these formulations for their two essential properties: their
ability to lower the surface or interfacial tension and their capacity to solubilize water-
insoluble compounds. As is shown in this volume, gemini surfactants are superior to
conventional surfactants in these two aspects. Another aspect in which gemini surfactants
could bring significant improvement is the flow behavior of a surfactant-containing
formulation. This aspect is crucial for many chemical engineering processes involving
mixing, pumping, or extruding. Rheological properties play an important role when
trying to satisfy consumer needs or to influence consumer preferences, and surfactants
have been long used as viscosity enhancers in personal care products. Currently, there are
some instances where they replace polymers (oil field, DNA analysis, etc.). The concept
of a gemini surfactant introduces new structural variables to control the rheology of
surfactant-based formulations.
This chapter reviews the rheology of isotropic micellar solutions of gemini and
oligomeric surfactants. It deals more specifically with solutions of cationic gemini
surfactants with a short spacer that form wormlike micelles, because those systems
present the most interesting rheological properties (see Fig. 2 of Chapter 1). Indeed, a
short spacer means that the two quaternary ammonium head groups in a gemini surfactant
are separated by a fixed distance shorter than the distance that would result from
thermodynamics in conventional surfactants. For symmetric gemini surfactants CmH2m+1
(CH3)2N+(CH2)sN+(CH3)2CmH2m+1, 2Br− (m and s are the carbonnumbers of the
hydrophobic tail and spacer) (m-s-m gemini surfactants) with m= 12, this means s<5 (see
Chapter 7, Section II). The average optimal surface area occupied by a short-spacer
gemini surfactant at the water-micelle interface being smaller, the spontaneous curvature
is reduced and happens to be just appropriate to induce the one-dimensional growth that
leads to wormlike micelles. The wormlike micelles can be sufficiently long and flexible
to act as entangled polymers, and this leads to a drastic increase of elasticity and
viscosity. Micellar growth and its rheological consequences are described in Section II,
along with some general considerations helpful for the remainder of the chapter. Section
III describes rheological properties of m-s-m surfactant solutions and points out some
peculiarities of their concentration dependence. The nonlinear viscoelastic properties of
gemini surfactant solutions are described in Section IV. Finally, the fifth section gathers
Gemini surfactants 184
some results about the rheology of gemini surfactant solutions in the presence of
additives. Although not so much studied up to now, these complex systems bear some
promising applications. In the course of this chapter, some rheological results pertaining
for asymmetrical quaternary ammonium gemini surfactants
[CmH2m+1(CH3)2N+(CH2)sN+(CH3)2 Cm′H2m′+1, 2Br−] referred to as m-s-m′, and oligomeric
quaternary ammonium surfactants [m-(s-m)z−1CH3(z=degree of oligomerization)] are also
reviewed.
II.
WORMLIKE MICELLES
A.
Growth of Wormlike Micelles
Micellization is a reversible association phenomenon. In the same way, wormlike
micelles are continuously breaking and merging. They are equilibrium polymers, which
means that their average length depends on concentration, temperature, and any other
thermodynamics state variables.
1.
Nonionic Surfactants
The early models of micellar growth, which neglected intermicellar interactions, account
well for the variation of the average micelle size with the surfactant concentration C
(expressed in weight fraction if not otherwise specified) and temperature T [10–13]. For
large micelles, the stepwise self-association model predicts an exponential equilibrium
distribution of micelle lengths:
Rheology of solutions of gemini surfactants 185
(1)
where C(L) is the concentration of micelles of length L. The average micelle length,
increases with concentration according to
(2)
where Ec is the end-cap energy. It corresponds to the excess energy associated to the
larger curvature in the two hemispherical end caps of a micelle. It is also twice the energy
necessary to break a wormlike micelle in two and is referred to as scission energy. In the
literature dealing with rheology, Ec is often expressed in kBT units (4× 10−21 J). For dilute
solutions of nonionic surfactants, Ec is independent of the aggregation number N. It is the
enthalpic driving force of growth of wormlike micelles. It is opposed to the entropy of
mixing of the whole system, which explains the broad distribution.
2.
Ionic Surfactants
For ionic surfactant micelles in salt-free solution, the scission energy contains an
electrostatic contribution that favors the breaking of the micelles. This contribution
decreases upon increasing concentration of the surfactant or adding salt, due to the
screening of the electrostatic interaction. This means that adding salt leads to an increase
of the micelle length. The ionic strength dependence of and Ec and the overall growth
of charged wormlike micelles, from very dilute to more concentrated regimes, have been
described theoretically [14–17]. Starting from a very low concen tration, three regimes
characterize the growth (see Fig. 1). At very low concentrations such that the Debye
length κ is longer than , the growth rate is low and the micelles are nearly spherical
and monodisperse. As the concentration increases, a sharp crossover to a rapid growth
regime occurs when becomes larger than κ. As in the case of neutral micelles, the
distribution is large, but the characteristic size grows faster than for neutral micelles,
according to
(3)
where lB is the Bjerrum length, a is the micelle diameter, and v is the apparent charge
density of the micelle.
Contrary to Eq. (2), Eq. (3) leads to a growth rate that is explicitly dependent on
concentration [17]:
(4)
In the third regime, at high concentration, the growth can be characterized by an effective
power law, with an exponent close to ½ as for neutral surfactant solutions.
Gemini surfactants 186
B.
Viscoelastic Behavior of Wormlike Micelles
1.
Rheology of Polymeric Systems
Rheology is the science of deformation and flow of matter. It aims at establishing and
understanding the mechanical response function of materials [18,19]. This is crucial for
many chemical engineering processes, as pointed out earlier.
Experimentally, the simplest strain is the simple shear. The sample is held between
two surfaces: one mobile and the other stationary separated by a small gap. When a step
strain γ is imposed to a viscoelastic liquid, the response in stress normalized by the strain
is the relaxation modulus G(t). It is illustrated schematically (Fig. 2) for an entangled
polymers solution, along with some molecular representation of the relaxation processes.
Note that very broad range of magnitude of G(t) and of time makes a log-log
Rheology of solutions of gemini surfactants 187
of strain dγ/dt approaches a limiting value given by σ/η0 and a steady state of flow is
reached. If one stops applying the stress (creep recovery), the elastic part of the
deformation is recovered. It is called the recoverable compliance J0. It is approximately
the reciprocal of G0.
Another possibility is to apply a sinusoidal strain at an angular frequency ω (rad/s). In
the linear regime, the resulting stress alternates also sinusoidally, but with a phase shift δ.
The stress can be decomposed in an in-phase component, the storage modulus G, and an
out-of-phase component, the loss modulus G″. The loss tangent is tan δ=G″/G′. The
frequency dependence of G′ and G″ are illustrated in Fig. 3, for a trimeric surfactant
solution,
2.
Structure, Molecular Weight, and Concentration Dependence of the
Viscoelasticity of Polymers
(a) Neutral Linear Polymers. Viscoelastic properties of polymers vary greatly with their
molecular weight and their concentration [18–20]. In the dilute regime, the dynamics of
polymers is described by the bead-spring models of Rouse and Zimm. Rouse’s model has
been extended to high concentration and melts of low-molecular-weight polymers. For
high-molecular-weight polymers, entanglements arise, which delay considerably the
complete relaxation. The main feature of the dynamics of entangled polymers is
described by the reptation model, which considers each polymer to be confined in a tube
[20,21]. To escape a strained tube and build a new one, polymers diffuse along their
backbone (reptation). The relaxation modulus in the terminal region is given by the
fraction of tube from which the polymer has not yet escaped. In other words, a segment
of strained tube is considered relaxed when it has been crossed by one end of the polymer
it contained (Fig. 4). For neutral entangled polymers, the concentration and contour
length (L) dependence of the characteristic rheological parameters are given by the
following scaling relations:
(5)
Experimentally, the viscosity and the relaxation time are found to scale as L3.4.
(b) Polyelectrolytes. The solution properties of charged polymers are very different
from those of uncharged ones. This has been observed for a long time and recently
theoretically accounted for [22,23]. An important feature of the semidilute behavior of
polyelectrolytes is the existence in a wide range of concentrations of a semidilute but not
entangled regime (C*<C<Ce) [22,23]. Ce/C* can be as large as 1000. This is because
polyelectrolytes interact with
each other from rather far, so the dilute-semidilute crossover is strongly shifted toward a
low concentration. In the semidilute, not entangled regime, the rheological properties are
expected to scale with molecular weight and concentration in a very peculiar way. The
relaxation time of polyelectrolytes decreases with concentration, the plateau modulus
varies linearly with concentration, and the viscosity follows Fuoss’ law:
(6)
At a higher concentration (C>Ce), when the chains overlap enough to entangle, the model
[22,23] predicts
(7)
At a still higher concentration, the dynamic properties cross over to those of uncharged
polymers [Eq. (7)].
(c) Branched Polymers. The viscoelastic behavior of classical polymers is drastically
modified when they are branched [24]. When the arms are not entangled, the viscosity
and the recoverable compliance are lower for branched polymers than for linear ones of
the same molecular weight. For entangled arms, the relaxation time is higher and the
relaxation spectrum broader. Reptation is hindered in branched polymers [20] and the
disentanglement process is dominated by the diffusion of the free end of the arms in a
potential centered at the junction. Each tube segment has its own characteristic relaxation
time, which depends on its distance from the junction. That explains the broadening of
the relaxation spectrum. However, the situation is completely different when wormlike
micelles are branched because the junction are fluid.
3.
Chemical Path of Relaxation in Wormlike Micelle Solutions
Figure 3 presents the rheological behavior of a 6% solution of 12–3–12–3–12. In the
frequency range corresponding to the terminal region of relaxation, it has the same
behavior as a mechanical Maxwell element (a spring and a dashpot combined in series).
This means that one single time characterizes the final relaxation process. At higher
frequencies, G′ increases slightly and G″ shows a pronounced minimum and the behavior
deviates from the maxwellian behavior.
This type of ideal viscoelastic behavior has been observed in many wormlike micelles
systems. The observation of a plateau region suggests a strong analogy with entangled
polymers. However, polymers with a very narrow distribution of molecular weight would
be required to observe such a terminal region. Actually, even a monodisperse polymer
sample would have a broader relaxation spectrum because of the various internal degrees
of freedom of the chain molecule. Reptation theory predicts for unbreakable
monodisperse polymers [25], whereas higher values are
observed experimentally.
The transient nature of the wormlike micelles and the role of reversible scission
explain the maxwellian behavior [26]. Micelles are continuously breaking and merging.
Rheology of solutions of gemini surfactants 191
than the one expected for rapid reversible scission and is well fitted by a stretched
exponential [45].
III
LINEAR VISCOELASTICITY OF AQUEOUS SOLUTIONS OF
GEMINI SURFACTANTS
A.
Nonmonotonic Concentration Dependence of the Zero Shear
Viscosity
The concentration dependence of the zero shear viscosity η0 of solutions of quaternary
ammonium oligomers m-(s-m)z−1CH3 [8,9,47] (see Fig. 8) and of asymmetrical dimeric
surfactants m-2-m′ [48,49] has been determined in a broad range of concentration. All
plots show a low concentration range where the viscosity of the solutions is very close to
that of water and increases slowly with C. At a certain concentration C*, the viscosity
increases rapidly with C, sometimes by several orders of magnitude, to reach a smooth
maximum at a concentration C**. The decrease of the viscosity at C>C** was rather
surprising when first observed on 12–2–12 [47]. A similar behavior had been previously
reported for a few conventional ionic surfactants such as the tetraethylammonium
perfluorooctylsulfonate [50]. This surfactant is one of the few that give rise to unscreened
charged wormlike micelles.
Rheology of solutions of gemini surfactants 195
The interpretation of C* obtained from viscosity data in terms of end-cap energy has to
be considered only qualitatively. It is now accepted that long wormlike micelles exist
even at concentrations below the value C* determined by viscosity measurements [41], as
was directly observed by cryo-TEM (transmission electron microscopy) [9,51].
The end-cap energy can, in principle, be estimated by fitting the concentration
dependence of the viscosity in the fast-growth regime. This requires a relation between
a b
η0, C, and the average micelle length of the form η0= C . The exponents a and b
have been shown to depend on the dominant relaxation process and on the stiffness of the
wormlike micelles [47]. The growth law L(C) is then introduced in this scaling relation to
obtain the C dependence of the viscosity of systems of wormlike micelles. In Ref. 11, the
very strong increase of viscosity above C* was supposed to result from the simultaneous
onset of the rapid micellar growth and of the crossover to the semidilute regime [47]. The
experimental data were found to lie between what would be expected for unbreakable
growing stiff cylinders and flexible polymers of short lifetime.
It has been suggested [8] that not all micelles are entangled in the concentration range
where the viscosity increases rapidly with C, because the scaling law for the elastic
modulus [see Eq. (7)] was not observed in this range. Other arguments supporting this
contention are given in Section III.B.1. Fuoss’ law [Eq. (6)] was used to fit the rapid
increases of viscosity [8]. This leads to reasonable values of the end-cap energy (40kBT
Rheology of solutions of gemini surfactants 197
for the dimer 12–3–12 and 80kBT for the trimer 12–3–12–3–12), but the fitting procedure
is not very robust and large errors are probably associated with these results.
B.
Nonuniversal Concentration Dependence of the Elastic Modulus
1.
Incipient Entanglement Behavior
Figure 3 shows that the storage modulus G′ reaches a plateau at high frequency. The
concentration dependence of the plateau modulus G0 of wormlike micelle solutions
follows the same scaling law as semidilute entangled polymer solutions. The magnitude
of the plateau modulus is rather insensitive to the chemical nature of the surfactant, and
the concentration dependence of G0 is probably the most universal feature of the
viscoelasticity of wormlike micelle solutions. This is expected because G0 depends on
neither the (not completely understood) relaxation mechanism nor the length of the
entangled polymer. It measures essentially the number of entanglement points ve per unit
volume in the transient network. By analogy with rubber elasticity, one can write:
G0=νekBT α C9/4 as predicted by Eq. (5). This scaling law is observed in many systems of
wormlike micelles [52]. For instance, data concerning cetyltrimethylammonium bromide
(CTAB) solutions at different salt concentrations merge onto the same universal curve
[8].
The plateau modulus of 12–2–12 aqueous solutions was reported to scale as C3, with a
slight dependence of the exponent on temperature [47]. The exponent is higher than that
expected for classical polymers, and the difference has been interpreted in terms of
electrostatic orientational correlations that reduce the elastic modulus and that are
stronger at low concentration [47].
Measurements of the elastic properties of 12–3–12 and 12–3–12–3–12 solutions in a
wider range of concentration revealed a more complex C dependence of the elasticity. At
low C, the experimental data lie below the universal curve and show a stronger
dependence on concentration (see Fig. 9). The irregular increase of the elastic properties
of the 12–3–12–3–12 solution has been attributed to the presence of branches and this
point is discussed in the next paragraph. However, the behavior of 12–3–12 was
surprising enough to deserve a new interpretation. It has been proposed that in this
concentration range, all of the wormlike micelles of the solution are not entangled [8,53].
The largest ones are probably entangled but not the shortest ones, as indicated by the
amplitude of the elastic modulus. The concentration Ce at which the elasticity data merge
the universal curve marks the onset of a fully entangled
Gemini surfactants 198
2.
Branching
Branching of threadlike micelles has been an intriguing problem for over a decade [58–
61] with particular attention to its consequences on rheology [62– 65]. The existence of
branched (connected) threadlike micelles was conjectured by Porte et al. [58]. The
existence of such micelles was also proposed in an attempt to explain the surprisingly low
viscosity of micellar solutions which had been shown by scattering techniques to contain
entangled threadlike micelles [62]. Abnormal behaviors noted in wormlike micelle
systems have been often attributed to branches. The trimeric surfactant 12–3–12–3–12,
which was shown by cryo-TEM to contain branches [66], provided the opportunity to
evidence the main consequences of branching for the rheology of the system. Elasticity
enhancement was observed, as expected in classical polymers, and analyzed to estimate
the branching density vb (see Fig. 12). Assuming additivity for the entanglement and
branching contributions to elasticity [67], one can write G0=(νb+νe)kBT. The
entanglement density was obtained from elastic modulus data on the 12–3–12 surfactant,
supposed to form strictly linear micelles. When looking at Fig. 9, elasticity enhancement
Rheology of solutions of gemini surfactants 201
is also clear in 12–2–12 solutions, where branching points had been suggested from
simulation [68] and were evidenced recently by cryo-TEM [69].
3.
Minimum in the Loss Tangent
In the plateau region, G′>G″ and the loss modulus goes through a minimum, G′ min (see
Fig. 3). The loss tangent defined as G″/G′ goes through a minimum also. On this time
scale, the Rouse motion of the strands between entanglements already proceeded, and
reptation did not yet significantly occur. For classical polymers, the minimum deepens as
the molecular weight increases [70] and slightly shifts toward low frequency. G″min/G0 is
directly related to the ratio of entanglement length over total chain length; le/ , and
should decrease upon increasing concentration [29,71]. The results in Fig. 13 show a
nonmonotonic variation of G″min/G0 for 12–2–12 and 12–3–12–3–12. A similar behavior
was also reported for inverse micelles of lecithin organogels [72] and interpreted as due
to branching. However, this hypothesis had been rejected for 12–2–12 [47], because
branching was supposed to lead to a weaker concentration dependence of the elastic
modulus. The nonmonotonic concentration dependence of the minimum loss tangent was
interpreted as a true decrease of micelle length. The temperature dependence of tan δmin
was also reported [47] and suggested that the end-cap energy decreases with
concentration.
Rheology of solutions of gemini surfactants 203
IV.
SHEAR THICKENING
When large strains or strain rates are allowed, the rheological behavior is extremely
diversified, shear fields being able to induce all kinds of structures, especially in
equilibrium structures like micelles. Nonlinear rheology of viscoelastic surfactant
solutions has been recently reviewed [73]. Shear thinning and shear thickening can be
observed. This section only reviews some shear thickening studies for 12–2–12 solutions
[41,74–77].
A.
Phenomenology
When 12–2–12 solutions are submitted to a constant shear stress and the resulting shear
rate γ is measured, the apparent viscosity is constant with shear rate up to a certain
critical shear rate γc, above which it increases. As γ is further increased, the viscosity
increases up to a maximum value and then decreases. The viscosity can be increased by a
factor of 10 at the maximum (Fig. 14a). Shear thickening is observed for solutions of 12–
2–12 for C between 0.05% and 1.8%, with a maximum amplitude around C*. γc
decreases with increasing C (Fig. 14b) or the addition of salt [41] and increases with
temperature [75]. These observations suggest that γc is related to the average micelle
length. It is noteworthy that the increase in viscosity happens after a certain latency time
(or equilibration time) that decreases with increasing shear rate. The latency time is also
very dependent on the thermal or shearing history of the solution [41].
B.
Molecular Interpretation
Rheological measurements of shear thickening have often been coupled with other
physical techniques of characterization such as flow birefringence, electrical
conductivity, cryo-TEM [75], small-angle neutron scattering (SANS) [77], or light
scattering [76]. It was shown that shear thickening is always associated with some
anisotropy in the sample and alignment of the micelles. After cessation of shear, the
birefringence decays in a time of the order of seconds (i.e., much shorter than the
recovery time of unperturbed rheological properties) [75]. SANS measurements showed
that the structural anisotropy actually persists over 100 s [77]. The same study evidenced
a small shift of the correlation peak to smaller values of the scattering vector, probably
indicating an aggregation phenomenon. Cryo-TEM pictures of sheared samples also
showed aggregation phenomenon [75]. Counterion-mediated aggregation has been
proposed to explain the large aggregates observed [77]. The first theoretical models
supposed the fusion of the end caps of aligned micelles, but they could not account for
the rather low γc observed and its salt dependence. A new mechanism has been proposed
in which the micelles collide under the shear flow and aggregate into bundles [78]. The
critical shear rate corresponds to the unbinding time. The bundles of micelles are then
supposed to form a network. A mechanism of aggregation involving the formation of
closed-loop micelles has also been proposed [79].
V.
EFFECT OF ADDITIVES
A.
Salt
The mechanical relaxation time of gemini surfactant solutions decreases upon the
addition of salt [47,80]. This has been also observed in many classical systems [81,82],
although not clearly understood. The surfactant 12–2–12 has been studied in the presence
of KBr, but this salt induces a phase transition to lamellar phases (see Chapter 9). More
recently, the same observation regarding the decrease of the relaxation time has been
made with NaF [41].
The rheology of 12–3–12 has been studied in the presence of 0.1 M KBr [83]. Figure
15a shows that in the presence of salt, C* decreases by an order of magnitude, but the
rate of increases of the viscosity with concentration is slower than in the absence of salt.
The elasticity data in Fig. 10 indicate a state of entanglement (G0 α C9/4), and the weight
averaged relaxation time decreases almost linearly with the concentration (Fig. 15b) [56].
Rheology of solutions of gemini surfactants 205
B.
Cosurfactant
The addition of hexanol to aqueous solutions of 12–2–12 leads to the formation of
vesicles [84]. From the point of view of rheological properties, this has the interesting
consequence of making the zero shear viscosity of the system less sensitive to
temperature, because when the temperature increases, vesicles transform into wormlike
micelles.
The rheology of mixed wormlike micelles of 12–2–12 and CTAB in 0.15 M KBr has
recently been studied as a function of the mole fraction R of 12–2–12
C.
Gemini-Polymer Systems
The rheology of mixtures of DTAB, 12–3–12 and 12–3–12–3–12 with hydrophobically
modified telechelic polymers (HMTP) consisting of polyethylene glycol, extended by
diisocyanates and end-capped with hexadecyl chains [85]. The addition of surfactant
brings about an increase of viscosity (see Fig. 16). However, this increase is larger with
12–3–12 and 12–3–12–3–12 than with the corresponding monomeric surfactant DTAB
and it is obtained at a lower content of surfactant. In addition, the viscosity increases over
the whole range of surfactant concentration due to the formation of wormlike micelles
[86]. The rheological behavior of the system is determined by the connectivity of the
surfactant-polymer mixed micelles, which strongly depends on the averaged functionality
of the micelles (it means the number of bridging HMTP that a micelle contains) [87,88].
On the other hand, the question may be raised as to whether the addition of polymers
to a system of gemini surfactant wormlike micelles brings any gain
VI.
CONCLUSION
Gemini surfactant solutions have interesting rheological properties due to their ability to
form wormlike micelles by virtue of their structure. They can even gel organic media,
and their interaction with polymers, although not well studied, offers interesting
perspectives. As compared to other wormlike micelle systems, gemini surfactant
solutions are free of salt and sometimes take original structures, such as branched
wormlike micelles or closed-loop micelles. The behavior of polyelectrolytes is still to be
understood, and gemini surfactant wormlike micelles offer good model systems for such
a goal. Nonlinear rheology of surfactant solution is also an active field of research, where
gemini surfactant are used extensively.
REFERENCES
21. Doi, M.; Edwards, S.F. The Theory of Polymer Dynamics. Oxford University Press: Oxford,
1986.
22. Rubinstein, M.; Colby, R.H.; Dobrynin, A.V. Phys. Rev. Lett. 1994, 73, 2776.
23. Dobrynin, A.V.; Colby, R.H.; Rubinstein, M. Macromolecules 1995, 28, 1859 and references
therein.
24. Grest, G.S.; Fetters, L.J.; Huang, J.S.; Richter, D. In Advances in Chemical Physics; Prigogine,
I., Rice, A., Eds.; Wiley: New York, 1996, Vol XCIV, 67.
25. Doi, M.; Edwards, S.F. The Theory of Polymer Dynamics’, Oxford University Press: Oxford,
1986;496 pp.
26. Löbl, M.; Thurn, H.; Hoffmann, H. Ber. Bunsenges. Phys. Chem. 1984, 88, 1102.
27. Cates, M.E. Macromolecules 1987, 20, 2289.
28. Turner, M.S.; Cates, M.E. Langmuir 1991, 7, 1590.
29. Granek, R.; Cates, M.E. J. Chem. Phys. 1992, 96, 4758.
30. Turner, M.S.; Cates, M.E. J. Phys. France 1992, 2, 503.
31. Shikata, T.; Hirata, H.; Kotaka, T. Langmuir 1987, 3, 1081.
32. Shikata, T.; Hirata, H.; Kotaka, T. Langmuir 1988, 4, 354.
33. Shikata, T.; Hirata, H.; Kotaka, T. Langmuir 1989, 5, 398.
34. Green, M.S.; Tobolsky, A.V. J. Chem. Phys. 1946, 14:80.
35. Yamamoto, M. J. Phys. Soc. Jpn. 1956, 11, 413.
36. Yamamoto, M. J. Phys. Soc. Jpn. 1957, 12, 1148.
37. Lodge, A.S. Trans. Faraday Soc. 1956, 52, 120.
38. Hoffmann, H.; Löbl, M.; Rehage, H.; Wunderlich, I. Tenside Detergents 1985, 22, 290.
39. Löbl, M.; Thurn, H.; Hoffmann, H. Ber. Bunsenges. Phys. Chem. 1984, 88, 1102.
40. Candau, S.J.; Merikhi, F.; Waton, G.; Lemaréchal, P. J. Phys. France 1990, 51, 977.
41. Oelschlaeger, C.; Waton, G.; Candau, S.J.; Cates, M.E. Langmuir 2002, 18, 7265.
42. Oelschlaeger, C.; Buhler, E.; Waton, G.; Candau, S. J. Eur. Phys. E 2003, 11, 7.
43. Cates, M.E. J. Phys. France 1988, 49, 1593.
44. In, M.; Aguerre-Chariol, O.; Zana, R. J. Phys. Chem. B 1999, 103, 7747.
45. Fischer, P.; Rehage, H.; Gruning, B. J. Phys. Chem. B 2002, 106, 11041.
46. Appell, J.; Porte, G. J. Phys. Lett. (Paris) 1983, 44, L-689.
47. Kern, F.; Lequeux, F.; Zana, R.; Candau, S.J. Langmuir 1994, 10, 1714.
48. Oda, R.; Huc, I.; Candau, S.J. Chem. Commun. 1997, 2105.
49. Oda, R.; Huc, I.; Homo, J.-C.; Heinrich, B.; Schmutz, M.; Candau, S. Langmuir 1999, 15, 2384.
50. Hoffmann, H. Adv. Colloid Interf. Sci. 1990, 32, 123.
51. Zana, R.; Talmon, Y. Nature 1993, 362, 228.
52. Hoffmann, H.; Rehage, H.; Rauscher, A. In Structure and Dynamics of Strongly Interacting
Colloids and Supramolecular Aggregates in Solution’, Chen, S-H.; Huang J.S.; Tartaglia, P.,
Eds.; Kluwer Academic: Boston, 1992; 493 pp.
53. In, M. Reactions and Synthesis in Surfactant Systems; Texter, J., Ed.; Marcel Dekker: New
York, 2001;59.
54. Schmitt, V.; Lequeux, F. J. Phys. II France 1995, 5, 193.
55. Massiera, G.; Ramos, L.; Ligoure, C. Europhys. Lett. 2002, 57, 127.
56. In, M. Unpublished data.
57. Kavassalis, T.A.; Noolandi, J. Macromolecules 1989, 22, 2709.
58. Porte, G.; Gomati, R.; El Haitamy, O.; Appell, J.; Marignan, J. J. Phys. Chem. 1986, 90, 5746.
59. Drye, T.J.; Cates, M. J. Chem. Phys. 1992, 96, 1367.
60. Elleuch, K.; Lequeux, F.; Pfeuty, P. J. Phys. I France 1995, 5, 465.
61. May, S.; Bohbot, Y.; Ben-Shaul, A. J. Phys. Chem. B 1997, 101, 8648.
62. Appel, J.; Porte, G.; Kathory, A.; Kern, F.; Candau, S.J. J. Phys. France 1992,2, 1045.
63. Khatory, A.; Lequeux, F.; Kern, F.; Candau, S.J. Langmuir 1993, 9, 1456.
64. Lin, Z. Langmuir 1995, 12, 1729.
65. Lequeux, F. Europhys. Lett. 1992, 19, 675.
Rheology of solutions of gemini surfactants 209
66. Danino, D.; Talmon, Y.; Levy, H.; Beinert, G.; Zana, R. Science 1995, 269, 1420.
67. Ferry, J.D. Viscoelastic Properties of Polymers, 3rd Ed.; Wiley: New York, 1980; 408.
68. Karaborni, S.; Esselink, K.; Hilbers, P.A.J.; Smit, B.; Karthäuser, J.; van Os, N.M.; Zana, R.
Nature 1994, 266, 254.
69. Berheim-Groswasser, A.; Zana, R.; Talmon, Y. J. Phys. Chem. B2000, 104, 4005.
70. Ferry, J.D. Viscoelastic Properties of Polymers, 3rd Ed.; Wiley: New York, 1980; 370.
71. Granek, R. Langmuir 1994, 10, 1627.
72. Shchipunov, Y.A.; Hoffmann, H. Langmuir 1998, 14, 6350.
73. Richtering, W. Curr. Opin. Colloid Interf. Sci. 2001, 6, 446.
74. Schmitt, V.; Schosseler, F.; Lequeux, F. Europhys. Lett. 1995, 30, 31.
75. Oda, R.; Panizza, P.; Schmutz, M.; Lequeux, F. Langmuir 1997, 13, 6407.
76. Weber, V.; Schosseler, F. Langmuir 2002, 18, 9705.
77. Oda, R.; Weber, V.; Lindner, P.; Pine, D.J.; Mendes, E.; Schosseler, F. Langmuir 2000, 16,
4859.
78. Barentin, C.; Liu, A.H. Europhys. Lett. 2001, 55, 432.
79. Cates, M.E.; Candau, S.J. Europhys. Lett. 2001, 55, 887.
80. Candau, S.J.; Hebraud, P.; Schmitt, V.; Lequeux, F.; Kern, F.; Zana, R. Nuovo Cimento 1994,
16D, 1401.
81. Khatory, A.; Kern, F.; Lequeux, F.; Appell, J.; Porte, G.; Morie, N.; Ott, A.; Urbach, W.
Langmuir 1993, 9, 933.
82. Khatory, A.; Kern, F.; Lequeux, F.; Candau, S.J. Langmuir 1993, 9, 1456.
83. In, M.; McCowen, S.A. Unpublished results.
84. Oda, R.; Bourdieu, L.; Schmutz, M. J. Phys. Chem. B 1997, 101, 5913.
85. Michaud, F.; In, M. Unpublished results.
86. Panmai, S.; Prud’home, R.K.; Peiffer, D.G. Colloids Surfaces A 1999, 147, 3.
87. Tanaka, F.; Edwards, S.F. J. Non-Newtonian Fluid Mech. 1992, 43, 247.
88. Annable, T.; Buscall, R.; Ettelaie, R.; Shepherd, P.; Whittlestone, D. Langmuir 1994, 10, 1060.
89. Guo, L.; Colby, R.H.; Lin, M.Y.; Dado, G.P. J. Rheol. 2001, 45, 1223.
9
Phase Behavior of Gemini Surfactants
RAOUL ZANA Institut C.Sadron (CNRS-ULP), Strasbourg, France
MARTIN IN CNRS-Université Montpellier 2, Montpellier, France
I.
INTRODUCTION
This chapter reviews the phase behavior of gemini (dimeric) surfactants. Phase behavior
is extremely important when it comes to properties and uses of surfactant-based
formulations. All manufacturers of personal care products, home care products, paints,
and coatings are interested in the phase behavior of the surfactant. It is important for
handling the final product itself as well as for the eventual processing in which the
surfactant is going to be involved. The information content in a phase diagram of a
surfactant is closely related to the expected behavior of a formulation in its various
conditions of application.
This is illustrated by quotations from the superb volume by Laughlin, which reviews
in depth and great detail the aqueous phase behavior of (conventional) surfactants [1].
The author states that “The evaporation of water from a liquid detergent (at the rim of a
bottle for example) may produce a liquid, a liquid crystal, or a crystal phase. Which is
formed depends on the intrinsic phase behavior of the product and the relative humidity”
[2]. The author further states,
This chapter focuses mainly on how the peculiar structure of gemini surfactants affects
their phase behavior and the phase diagram of gemini surfactant-water mixtures with
respect to the corresponding monomeric surfactant-water mixtures. Published studies
devoted to phase diagram of gemini-surfactant-containing systems are scarce. Indeed, as
stated by Laughlin [3], “relevant phase diagrams are highly valued, but few people are
interested in their determination.” This chapter provides an opportunity to review
published work and to present some unpublished results. In a review of the phase
behavior of surfactants, one should, in principle, be concerned with all of the
thermodynamic information available in the literature. Because the solubility in water and
Phase behavior of gemini surfactants 211
II.
THERMOTROPIC BEHAVIOR OF GEMINI SURFACTANTS
directly yielded an isotropic surfactant melt with surfactants 16–1–16 and 16–2–16,
whereas surfactants 16–3–16 and 16–6–16 showed a lamellar-to-smectic A (lamellar Lα
phase) transition at around 200–230 °C before transforming into isotropic surfactant melt.
Contrary to 12-s-12 surfactants, the 12-EOx-12 surfactants displayed thermotropism
[7], similar to the 16-s-16 surfactants. An important difference is the softening transition
temperature that occurs at a lower temperature (about 32 °C). The surfactants with x ≤ 4
showed the viscous neat and smectic A phases with transition temperatures dependent on
the spacer length. The layer spacing did not vary much with the spacer length. This
difference of behavior between 12-s-12 and 12-EOx-12 surfactants emphasizes the
importance of the nature of the spacer, in addition to its length, on the properties and
behavior of gemini surfactants. The conformations allowed to the spacer can vary
substantially depending on its nature. In particular, the presence of the oxygen atom in
the EOx spacer increases its flexibility with respect to a polymethylene spacer. This, in
turn, may affect the packing of the head groups and thus the phase behavior of the
surfactant.
III.
LYOTROPISM IN GEMINI SURFACTANT-WATER MIXTURES
A.
Theoretical Considerations on the Aqueous-Phase Behavior of
Surfactants
The following considerations are largely based on the works by Israelachvili et al. [12]
and Mitchell et al. [13]. Only the essential aspects of these articles that are relevant to
surfactant phase behavior have been retained and simplified as to give the reader some
markers in the understanding of phase behavior.
Surfactants self-associate in water, giving rise to micelles where the alkyl chains are
bundled together forming an oily core. The core is coated by the surfactant polar head
groups, which largely prevent contacts between water and surfactant alkyl chains. The
formation of micelles is a cooperative process that involves a large number of surfactants
(typically 20–100) and that occurs in a relatively narrow range of surfactant
concentration, the critical micellization concentration (cmc). Micelle formation can be
viewed as a chemical equilibrium whereby a certain number of surfactants associate
reversibly to form a micelle. In the case of ionic surfactants, part of the surfactant
counterions contribute to the formation of the micelles. Nevertheless, the resulting
Gemini surfactants 214
micelles are charged and the micelle ionization degree is in the range 0.15–0.3 for usual
ionic surfactants.
The driving force for surfactant self-assembly into micelles or vesicles is the so-called
hydrophobic (attractive) interaction between surfactant alkyl chains. The origin of this
interaction lies in the entropically unfavorable structure adopted by the water molecules
surrounding alkyl chains with respect to bulk water. When alkyl chains are immersed into
water, the system will tend to minimize its free energy by eliminating contacts between
alkyl chains and water molecules. This is achieved by having the alkyl chains “contacting
each other.” With surfactants, macroscopic phase separation is avoided due to competing
repulsive interactions at the head-group level. The main forces opposing self-association
are repulsive electrostatic forces between polar or charged head groups, steric forces
involved in the packing of the alkyl chains in the micelle core and of the head-groups at
the micelle surface, and a force arising from the residual contacts between alkyl chains
and water at the micelle surface. The balance between attractive and repulsive forces
results in the formation of micelles of finite size at low concentration, close to the cmc. It
also determines the shape of the micelles.
At the cmc, the micelles are generally spherical or spheroidal. The radius of the
micelle core can be taken as the length l of the all-trans form of the surfactant alkyl chain.
As the surfactant concentration is increased, the micelles may remain spherical or grow
and undergo a change of shape. In dilute solutions, this will be determined by the
structural characteristics of the surfactant (i.e., the length l and volume v of the surfactant
alkyl chain and the optimal value of the surface area a occupied by one head group at the
micelle surface). The optimal value of a is that which minimizes the free energy of the
system. The three quantities l, v, and a are lumped together in the so-called surfactant
packing parameter P=v/al [12]. This parameter determines the curvature of surfactant
aggregates. Surfactants with P≤⅓ form spherical micelles which remain spherical upon
increasing surfactant concentration. Surfactants with ⅓<P<½ give rise to spherical
micelles at the cmc, but these micelles grow with concentration into elongated micelles.
The closer P is to ½, the steeper the micelle growth. Micelles of surfactants with ½≤P<1
grow from spherical micelles to disk micelles as the concentration increases. Finally, for
P>1, the surfactant forms reverse aggregates where the water is contained in volumes
having the shape of spheres, cylinders or disks.
So far, only intramicellar interactions have been considered. Those forces work
parallel to the interface. As the surfactant concentration increases further, intermicellar
forces come into play. They are repulsive and work perpendicular to the micelle-water
interface [14]. Repulsive interactions induce ordering with symmetry and limiting
volume fraction Φmax that depend on the shape of the micelles. Spherical micelles close
pack into a cubic array (with Φmax=0.74), rodlike micelles form hexagonal (biaxial) array
(Φmax = 0.92), and bilayers form lamellar phase (Φmax=1.00) [14]. The most important
point is that this simple approach leads one to predict the following sequence of phases
for a surfactant-water mixture if the surfactant forms spherical micelles (P<⅓):
Spherical micelles → Cubic phase → Hexagonal phase → Lamellar phase
If the surfactant used can form elongated or disk micelles then the sequences of phases
are, respectively, as follows:
Disordered elongated micelles → Hexagonal phase → Lamellar phase
Phase behavior of gemini surfactants 215
However, the intermicellar interactions also include a soft-core repulsion. This additional
interaction can be relatively long range and can assist the formation of cubic structures
other than of the face-centered type [13]. The unfavorable energy curvature can then be
compensated by a contribution from intermicellar interactions. Because intermicellar
interactions work perpendicular to the micelle-water interface, any decrease in the
distance between micelles (upon increasing concentration) increases the free energy of
the system. A way for the system to compensate this effect and to maintain an
intermicellar distance as large as possible is to reduce the number of micelles, (i.e., to
cross over from spherical to cylindrical micelles or from cylinders to sheets) [14]. That
makes possible the occurrence of the transitions:
Spherical micelles → Hexagonal phase → Lamellar phase
Disordered elongated micelles → Lamellar phase
B.
Effect of the Spacer Length and Nature
An x-ray study of 12-s-12-water mixtures showed that lyotropic mesophases occurred in
a concentration range which became more narrow as s increased and completely
disappeared for s=10 and 12 [6]. For these two surfactants, the micellar range extends to
concentrations as high as 90 wt%, a behavior similar to that reported for the
corresponding monomeric surfactants 12–5 and 12–6 [15] and other cationic surfactants
with a bulky headgroup such as dodecyltributylammonium chloride [16]. The reason for
the absence of a mesophase lies mainly in the difficulty for the micelles to change their
shape as a result of a very strong packing constraint. The formation of prolate
polydisperse micelles could thus explain the absence of ordering in gemini 12–12–12 as
well as in the corresponding surfactant monomer 12–6. Lyotropic mesophases were again
observed for s≥16, having the texture of the conventional lamellar and cylindrical
(hexagonal) phases. The alkyl chains were located inside the cylinders in the hexagonal
phase. The spacer group lays nearly flat and almost fully extended on the core-water
interface in both the lamellar and cylindrical phases, up to s=8 [6]. A Raman
spectroscopy study of m-6-m surfactants confirmed this result [17]. The phase diagram of
the 12–8–12 surfactant given in Fig. 2 illustrates the phase behavior of 12-s-12
surfactants. The absence of the cubic phase is surprising because the 12–8–12 micelles
are known to be spherical even at a relatively high surfactant concentration (see Fig. 3 of
Chapter 7). This illustrates the importance of intermicellar interactions on the shape of
the micelles at
C.
Effect of the Alkyl Chain Length and Nature
It has been shown in Section III.B that the 16-s-16-water mixtures [11] show a richer
lyotropism than the 12-s-12-water mixtures [6] with, in particular, a bicontinuous cubic
phase and an intermediate phase that are not observed with the second type of mixture.
The phase diagrams of the surfactants 8–6–8, 10–6–10 and 12–6–12 have been
mapped by DSC. They are represented in Fig. 3 [17]. The authors distinguish five
regions: region I corresponds to the micellar L1. Region II and III were called
respectively gel phase and coagel. The region IV corresponds to the coexistence of
phases II and III. No information was provided on the symmetry of those structures. The
fifth region is constituted by hydrated crystals. Polarized light microscopy suggests that
region II corresponds to the H1 phase [22]. The increase of the alkyl chain length is seen
to bring about a progressive decrease of the realm of the micellar phase I and a
progressive increase of the range of the “gel” phase II. It also brings about the occurrence
of the coagel phase III at progressively lower concentrations and the occurrence of region
V for the 12–6–12 surfactant, where hydrated crystals of surfactant coexist with an
aqueous solution. The Raman study of the systems suggested the stabilization of the all-
trans form of the alkyl chain and of the spacer in the gel phase and particularly in the
coagel phase. As with conventional cationic surfactants, mesophases are more robust for
m=16 than m=12.
The partial phase behavior of dissymmetric gemini surfactants m-2-m′ has been
reported [23]. Freeze-fracture and x-ray scattering showed that the surfactant 8–2–16
forms stacked bilayers with very little water between bilayers and an inverted hexagonal
phase at higher surfactant concentration. The surfactant 14–2–18 formed ribbons at a
concentration of only 3 wt%. Increasing concentration led to multibilayer ribbons and
ordered smectic domains. The authors emphasized that the m-2-m′ surfactants never gave
rise to the normal lamellar phase where the interlamellar spacing increases with dilution.
The aqueous phase behavior of the sugar-based surfactants (CmH2m+1)2
C[CH2NHCO(CHOH)4CH2OH]2 has been determined by the solvent penetration optical
microscopy method and deuterium NMR spectroscopy [24,25]. The surfactants with m=5
and 6 showed, at around 25°C, the phase sequence L1, H1, V1, Lα upon increasing
concentration (Figs. 4a and 4b). The cubic phase V1 disappeared with the m=7 surfactant
(Fig. 4c). The surfactant with m=8 showed only the L1 and Lα phases (Fig. 4d). However,
a phase denoted “X” was present in a narrow range of concentration [25]. This
Phase behavior of gemini surfactants 219
shape of the micelles (cylindrical for m=5 and 6 and disklike for m=8 on the basis of
small-angle neutron scattering experiments).
D.
Comparison of the Aqueous-Phase Behavior of Monomeric and
Dimeric or Oligomeric Surfactants
The phase diagrams of some conventional surfactants and the corresponding dimeric or
oligomeric surfactants have been determined in only a few instances. This is not
surprising in view of the effort involved in the determination of a phase diagram.
N-dodecanoyl-N-methyl β-alanate
(MS) at 25°C. Wm: aqueous micellar
solution phase; S: solid; II: region
containing two types of liquid crystals.
(Reproduced from Ref. 28 with
permission of the American Chemical
Society.)
The phase diagrams of the gemini surfactant [Na+,C11H23C(O)N(CH2)2 CO2−]2(CH2)2
and of the corresponding conventional surfactant Na+,C11 H23C(O)N(CH3)(CH2)2CO2− are
represented in Fig. 5 [28]. These phase diagrams are strikingly similar and again show the
sequence L1/H1/V1/Lα, as predicted in Fig. 1. The main difference is that the cubic phase
V1 occurs at a much lower temperature for the monomeric than for the dimeric
surfactant. The absence of the micellar cubic phase is not surprising for the gemini
surfactant, as the spacer group is rather short and, thus, the micelles are probably
cylindrical at high concentration, close to the micellar/hexagonal boundary line.
Figure 6 compares the phase diagrams of nonionic surfactants: the conventional
surfactant Triton X-100 [p-tert-octylphenylene(EO)9.5] [29] and the oligomeric surfactant
Tyloxapol [30,31]- In fact, Tyloxapol is a mixture of oligomeric surfactants with a repeat
unit very close to Triton X-100, a maximum degree of polymerization of 7, and
methylene groups connecting the phenylene moieties of the repeat units. The similarity
between the phase diagrams is as striking as in the preceding example. For both
surfactants, the phase of largest extension is the hexagonal H1 phase that immediately
follows the micellar phase. Both surfactants show the usual sequence L1/H1/V1/Lα, but the
mesophases melt at temperatures about 30 °C lower for the conventional surfactant. In
addition, the lamellar phase of Tyloxapol apparently contained small droplets of another
phase that may be a Tyloxapol hydrate [30].
E.
Lyotropic Mesophase Transformation Induced by Additives
The thread-like micellar phase of 12–2–12 (surfactant concentration above 5 wt%) is
transformed into a lamellar phase upon addition of KBr [33]. Further addition of KBr
results in phase separation into a nearly pure KBr solution (lower phase) and a
birefringent KBr-poor, surfactant-rich lamellar upper phase. A study of this phase by
small-angle neutron scattering (SANS) showed a dramatic increase of order upon
increasing the temperature to 30 °C, which was first attributed to a Lα/Lβ, phase
transition. The study also revealed the presence of highly curved defects that were
identified as water-filled holes in the lamellae. Later, more detailed SANS studies [34]
revealed the existence of two lamellar phases. The first one is metastable, could be
diluted with water, and corresponds to the Lα phase identified in the earlier study [34].
The existence of water-filled holes is no longer certain and that phase may, in fact, be a
mixture of an Lα phase and an isotropic phase. Over time, this phase transforms into the
second lamellar phase that is characterized by a constant swelling and an interlamellar
distance of about 7 nm. This phase contains relatively little water. These results are
similar to those for the dissymmetric gemini surfactants m-2-m′ that can form stacks of
bilayers with very little water between bilayers (see Section III.C) [23,35]. Such a
behavior is indicative of strong attractions between bilayers of gemini surfactants with a
short spacer. The bromide ions may act as bridges between oppositely charged interfaces
[23]. This explanation is similar to that for the existence of two lamellar phase in the
mixtures of the gemini surfactants [Br−, N≡C—C6H4—C6H4O(CH2)mN+(CH3)2]2(CH2)6
with water [11].
The addition of hexanol to a solution of wormlike micelles of the 12–2–12 surfactant
was found to result in the formation of vesicles [36]. This transition was attributed to the
decrease of spontaneous curvature of the surfactant film upon incorporation of hexanol.
Gemini surfactants 224
IV.
PHASE BEHAVIOR OF GEMINI SURFACTANTS IN THE
PRESENCE OF OIL AND COSURFACTANT
V.
MISCELLANEOUS STUDIES
This section reviews studies that involve, in part, the phase behavior of gemini
surfactants. Gemini surfactants 16-s-16,2C15H31CO2−, where the counterion is a palmitate
Gemini surfactants 228
anion and s=2–12, have been reported [42]. Their investigation by DSC revealed a gel-to-
liquid crystalline (lamellar) phase transition with a transition temperature that decreased
rather rapidly for s between 3 and 4 and
VI.
CONCLUSIONS
REFERENCES
1. Laughlin, R.G. The Phase Aqueous Behavior of Surfactants. Academic Press: London, 1994.
2. Laughlin, R.G. in Ref. [1], page 9.
3. Laughlin, R.G. Curr. Opin. Colloid Interface Sci. 2001, 6, 405.
4. Skoulios, A. Ann. Phys. 1978, 3, 421.
5. Laughlin, R.G. in Ref. [1], (a) pages 226–227; (b) pages 351–352.
6. Alami, E.; Levy, H.; Zana, R.; Skoulios, A. Langmuir 1993, 9, 940.
7. Dreja, M.; Gramberg, S.; Tieke, B. Chem. Commun. 1998, 1371.
8. Zana, R. J. Colloid Interface Sci. 2002, 252, 259.
9. Alami, E.; Levy, H.; Zana, R.; Weber, P.; Skoulios, A. Liq. Crystals 1993, 13, 201.
10. Fuller, S.; Hopwood, J.; Rahman, A.; Shinde, N.; Tiddy, G.J.; Attard, G.S.; Howell, O.;
Sproston, S. Liq. Crystals 1992, 12, 521.
11. Fuller, S.; Shinde, N.; Tiddy, G.J.; Attard, G.S.; Howell, O. Langmuir 1996, 12, 1117.
12. Israelachvili, J.N.; Mitchell, D.J.; Ninham, B.W. J. Chem. Soc. Faraday Trans. 2 1976, 72,
1525.
13. Mitchell, D.J.; Tiddy, G.T.; Waring, L.; Bostock, T.; McDonald, M.P. J. Chem. Soc. Faraday
Trans. 1983, 79, 975.
14. Charvolin, J.; Sadoc, J.-F. In Micelles, Membranes, Microemulsions, and Monolayers; Gelbart,
W.M., Ben-Shaul, A. Roux, D., Eds.; Springer-Verlag: New York, 1994.
15. Hertel, G.; Hoffmann, H. Prog. Colloid Polym. Sci. 1988, 76, 123.
16. Blackmore, E.S.; Tiddy, G.J.T. J. Chem. Soc. Faraday Trans. 2 1988, 84, 1115.
17. Hattori, N.; Hara, M.; Okabayashi, H.; O’Connor, C.J. Colloid Polym. Sci. 1999, 277, 306.
18. Pestman, J.; Terpstra, K.R.; Stuart, M.C.; van Doren, H.A.; Brisson, A.; Kellogg, R.M.;
Engberts, J.B.F.N. Langmuir 1997, 13, 6857.
19. Pérez, L.; Torres, J.L.; Manresa, A.; Solans, C.; Infante, M.R. Langmuir, 1996, 12, 5296.
20. Laughlin, R.G. In Cationic Surfactants; Jungermann, E. Ed., Marcel Dekker: New York, 1970;
p. 1.
21. Pinazo, A.; Diz, M.; Solans, C.; Pes, M.A.; Erra, P.; Infante, M.R. J. Am. Oil Chem. Soc. 1993,
70, 37.
22. M. In, unpublished results.
Gemini surfactants 230
23. Oda, R.; Huc, I.; Homo, J.C.; Heinrich, B.; Schmutz, M.; Candau, S.J. Langmuir 1999, 15,
2384.
24. Eastoe, J.; Rogueda, P.; Harrison, B.J.; Howe, A.M.; Pitt, A.R. Langmuir, 1994, 10, 4429.
25. Eastoe, J.; Rogueda, P.; Howe, A.M.; Pratt, A.R.; Heenan, R.K. Langmuir 1996, 12, 2701.
26. Gazeau, D.; Zemb, Th.; Dubois, M. Prog. Colloid Polym. Sci. 1993, 93, 123.
27. Ockelford, J.; Timimi, B.A.; Narayan, K.S.; Tiddy, G.T. J. Phys. Chem. 1993, 97, 6767.
28. Kunieda, H.; Masuda, N.; Tsubone, K. Langmuir 2000, 16, 6438.
29. Heusch, R. Tenside Surfactants Detergents 1991, 28, 38.
30. Westesen, K. Int. J. Pharm. 1994, 102, 91.
31. Westesen, K.; Koch, M.H. Int. J. Pharm. 1994, 103, 225.
32. Laschewsky, A. Adv. Polym. Sci. 1995, 124, 1 (and references therein).
33. Buhler, E.; Mendes, E.; Boltenhagen, P.; Munch, J.P.; Zana, R.; Candau, S.J. Langmuir 1997,
13, 3096.
34. Knaebel, A.; Oda, R.; Mendes, E.; Candau, S.J. Langmuir 2000, 16, 2489.
35. Oda, R.; Huc, Y.; Candau, S.J. Chem. Commun. 1997, 2105.
36. Oda, R.; Bourdieu, L.; Schmutz, M. J. Phys. Chem. B 1997, 101, 5913.
37. Dreja, M.; Tieke, B. Langmuir 1998, 14, 800.
38. Dreja, M.; Pickhout-Hintzen, W.; Mays, H.; Tieke, B. Langmuir 1999, 15, 391.
39. Dreja, M.; Tieke, B. Ber. Bunsenges. Phys. Chem. 1998, 102, 1705.
40. Perez-Luna, V.; Puig, J.; Castano, V.; Rodriguez, B.; Murthy, A.; Kaler, E. Langmuir 1990, 6,
1040.
41. Layn, K.L.; Debenedetti, P.G.; Prud’homme, R.K. J. Chem. Phys. 1998, 109, 5651.
42. Bhattacharya, S.; De, S. Langmuir 1999, 15, 3400.
43. van Doren, H.A.; Smith, E.; Petsman, J.M.; Engberts, J.B.F.N.; Kellogg, R.M. Chem. Soc. Rev.
2000, 29, 183.
44. Sommerdijk, N.A.; Lambermon, M.H.; Feiters, M.C.; Nolte, R.J.; Zwanenburg, B. Chem.
Commun. 1997, 1423.
45. Oda, R.; Huc, I.; Candau, S.J. Angew. Chem. Int. Ed. 1998, 37, 2689.
46. Menger, F.M.; Peresypskin, A.V.; Caran, K.L.; Apkarian, R.P. Langmuir 2000, 16, 9113.
47. Strey, R.; Jahn, W.; Porte, G.; Bassereau, P. Langmuir 1990, 6, 1635.
10
Mixed Micellization Between Dimeric
(Gemini) Surfactants and Conventional
Surfactants
RAOUL ZANA Institut C.Sadron (CNRS-ULP), Strasbourg, France
JIDING XIA Wuxi University of Light Industry, Wuxi, People’s
Republic of China
I.
INTRODUCTION
would make the use of gemini surfactants even more attractive than from the simple fact
that their cmc is much lower than for the corresponding conventional surfactants. The
second type of studies were performed for the sake of understanding mixed micellization
in mixtures of conventional and gemini surfactants, measuring cmc and micelle size, and
investigating the microstructure of the solutions.
Studies in relation to synergism were essentially performed using surface tension
measurements, both static and dynamic. The physico-chemical investigations of mixtures
of surfactants containing gemini surfactants used electrical conductivity and pyrene
fluorescence probing [5] (measurements of cmc), as well as neutron [6] and light [7]
scattering and time-resolved fluorescence quenching (TRFQ) [5] (measurement of
micelle size or aggregation number). Transmission electron microscopy at cryogenic
temperature (cyro-TEM) [8] was used for the direct visualization of the mixed-solution
microstructure.
Many of the studies reviewed in this chapter were performed with the gemini
surfactants alkanediyl-α,ω-bis(alkyldimethylammonium bromide) that are referred to as
m-s-m, s and m being the carbon numbers of the alkanediyl spacer and alkyl chain,
respectively.
II.
STUDIES OF SYNERGISM IN MIXED CONVENTIONAL
SURFACTANT-GEMINI SURFACTANT SYSTEMS
A.
Synergism in Surfactant Mixtures
In the case of mixed-surfactant systems, the word synergism is used when the properties
of the mixtures are superior to those of the isolated surfactants. Rosen et al. [9–11] have
discussed the occurrence of synergism in the formation of mixed micelles, the surface
tension reduction efficiency, and the surface tension reduction effectiveness on the basis
of nonideal solution theory. The efficiency in surface tension reduction is characterized
by the surfactant concentration C20 (see above) [11]. The surface tension reduction
effectiveness can be defined as the reduction of surface tension attained at the cmc of the
system [11]. It corresponds to the maximum lowering of surface tension that can be
obtained by a given surfactant or surfactant mixture because the surface tension hardly
changes at concentrations above the cmc. A surfactant mixture is said to show synergism
when the values of the cmc, C20, or the surface tension at the cmc are for some mixture
composition lower than for the isolated surfactants making up the mixture. The search
for, and understanding of, synergism in surfactant mixtures is important when it comes to
applications and uses of surfactants.
The conditions for the occurrence of synergism have been determined. For a binary
mixture of surfactants 1 and 2, the conditions for synergism in mixed micellization are
[9]
(1)
Mixed micellization between dimeric mixed micellization between dimeric surfactants 233
βm is the molecular interaction parameter between surfactants 1 and 2 in the bulk phase
and and are their respective cmc.
Likewise, the conditions for synergism in surface tension reduction efficiency are [9]
(2)
(3)
γ°cmc1 and γ°cmc2 are the values of the surface tension at the cmc for surfactants 1 and 2. K
is the slope of the γ-ln C plot of the individual surfactant having the larger surface tension
at its cmc.
At the outset, it is important to stress that the values of the molecular interaction
parameters depend heavily on the nature of the head groups because the vast majority of
the investigated surfactants have alkyl chains as hydrophobic moieties. Nevertheless, they
also depend on the length of the surfactant alkyl chain, as will be seen below,
emphasizing the importance of chain packing on the properties of mixed systems.
B.
Review of the Reported Results
Most of the reported studies in relation to synergism have been performed by Rosen et al.
[12–16], using surface tension to determine the cmc of various mixtures and fitting
appropriate equations to the data to obtain the molecular interaction parameters βm and/or
βσ. Additional studies determined values of cmc by electrical conductivity or
fluorescence probing and reported only βm values [17–20]. The more negative the values
of βm and βσ, the more likely the existence of synergism in the surfactant mixture. Rosen
et al. have devised a method that permits the determination of the molecular interaction
parameters from the surface tension versus concentration plots for the two isolated
surfactants and at least one binary mixture of the two surfactants.
Figures 1–3 illustrate the results obtained on mixed gemini surfactant-conventional
surfactant systems from which information on synergism is extracted. Figure 1 shows the
surface tension versus concentration plots (log scale) for a system where the synergism in
micellization and in surface tension reduction efficiency is clearly seen. Indeed, the plot
for the mixture shows a cmc and a C20 value that are smaller than for the isolated
surfactants. Figure 2 corresponds to a mixture where there is synergism in micelle
formation and in
Gemini surfactants 234
41 12–2-C4H8C8F17—C8F17C4H8–2-C4H8C8F17 −0.84 22
a
Abbreviations used: The chemical structures of most gemini surfactants are given under the form
(amphiphilic moiety)2 spacer, except C10DADS: C10H21C6H2(C10H21)(SO3−Na+)OC6H4 SO3−Na+.
C10OC10: {C10H21OCH2CH[O(CH2)3SO3−Na+]}2CH2OCH2
C10E3C10: {C10H21OCH2CH[O(CH2)3SO3−Na+]}2(CH2OCH2)4
C8C1C8: {C8H17OCH2CH[O(CH2)3SO3−Na+]}2+]}2CH2OCH(CH2OCH3)CH2OCH2
Dim1: (C10H21CHOSO3−Na+)2(CH2OCH2)3
Dim3: [C7H15CH(SO3−Na)C(O)O]2(CH2)2
Dim4: [C10H21OC(O)CH(SO3−Na+)]2(CH2)6
m-s-m: [CmH2m+1(CH3)2N+]2(CH2)s
m-spacer-m: [CmH2m+1(CH3)2N+]2 spacer
C14DMAO: tetradecyldimethylamine oxide
C12E5, C12E6, C12E7, C12E8: penta-, hexa-, hepta-, and octaethyleneglycol monododecylether
SDeSO3, SDSO3: sodium decyl and dodecyl sulfonate
SDS: sodium dodecylsulfate
SDE4SO4: sodium dodecyltetraethyleneglycol sulfate
C10Maltoside, C12Maltoside: decyl- and dodecylmaltoside
DeTAB, DTAB, CTAB: decyl-, dodecyl-, and hexadecyltrimethylammonium bromide
surface tension reduction efficiency and effectiveness. Figure 3 shows the variations of
the cmc with the mixture composition for two systems that show synergism in micelle
formation.
Table 1 list values of βm and βσ for for gemini surfactant-conventional surfactant
mixtures and also for some of the corresponding conventional surfactant-conventional
surfactant mixtures (indicated by the entries in bold) [13–15,17–21]. The following
trends are apparent:
1. The values of βm and βσ are are generally less negative for the gemini surfactant-
conventional surfactant mixtures than for the corresponding conventional surfactant-
conventional surfactant mixtures. Compare, for instance, entries 1–4 to entry 5, entries
11 and 12 to entry 13, entries 14 and 15 to entry 16, entries 29 and 30 to entry 31, and
entry 32 to entry 33. Rosen et al. [12] attributed this behavior to a steric hindrance in
the interaction between the two surfactants, due to the peculiar chemical structure
(bulkiness) of gemini surfactants. Nevertheless, the cationic gemini surfactant-sugar-
based conventional surfactant mixtures (entries 22–25 and 27) show an opposite
behavior with more negative values of βm and βσ than the corresponding conventional
surfactant-sugar-based surfactant mixtures (entries 26 and 28) [15]. This was
explained by the fact that the gemini surfactant has two positively charged head
groups interacting with the weakly negatively charged sugar head groups at pH=9
used in the experiments, whereas the conventional surfactant has only one positively
charged head group.
2. In all instances where both molecular parameters were determined, the value of βσ is is
more negative than that for βm. That behavior indicates that the interaction between
surfactants is stronger in flat monolayers than in strongly curved micelles. It reflects
the greater difficulty of incorporating the two hydrophobic groups of a gemini
surfactant into a spherical micelle than into a planar air-water interface [23].
Mixed micellization between dimeric mixed micellization between dimeric surfactants 239
3. As would be expected, the values of βm and βσ for for ionic gemini surfactant-
oppositely charged conventional surfactant mixtures are very large (entries 17–20).
Similarly large values of βm and βσ have have been reported for mixtures of oppositely
charged conventional surfactants [21]. The peculiar behavior of mixture entry 21 has
been attributed to the formation of a soluble complex that is not surface active in a
molar ratio 1:1 of the two surfactants [14].
4. The effect of the length of the surfactant appears to be important. For instance, in
entries 35–37, the length of the gemini surfactant alkyl chain is increased while the
conventional surfactant remains unchanged. The systems show a considerable increase
of |βm|, indicating a stronger tendency to comicellization and to synergism as the
lengths of the alkyl chains of the two surfactants tend to become equal. It would have
been interesting to check whether |βm| decreases when the gemini alkyl chain becomes
longer than that of the conventional surfactant. A similar although less important
effect is seen in entries 34 and 35 when the alkyl chain length of the conventional
surfactant is increased while that of the gemini remains unchanged.
5. The effect of the nature of the gemini surfactant spacer group appears to be not so
important. For instance, entries 22 and 24 that correspond to gemini surfactants with
spacer groups of equal length but different hydrophobicity show nearly equal values of
both βm and βσ.
6. The length of the spacer group of the gemini surfactant appears to also have little effect
on the values of βm and βσ (compare entries 3 and 4, or 8 and 9, or 22–25, or 29 and
30). Nevertheless, a study of a series of gemini surfactant-sugar-based conventional
surfactant mixtures showed that the gemini surfactant having the lowest surface area at
the air-water interface also shows the largest value of |βσ| and the lowest value of |βm|
[15].
Rosen et al. explained the changes in values of βm and βσ in in terms of steric hindrance at
the level of the head group of the gemini surfactants [12,13], changes of surface area per
surfactant at the air-solution interface [15], protonation of ether oxygen atoms in the
surfactant spacer or head group [13], and complexation of the surfactant counterion (Na+)
by poly(ethylene oxide) head groups [16], depending on the mixed system investigated.
The values of βm and βσ together with those of cmc1 and cmc2, and [see Eq. (2)],
K, and γ°cmc1 and γ°cmc2 [see Eq. (3)] have been used to discuss the occurrence of
synergism in micellization and in surface tension reduction efficiency and effectiveness.
The results suggest that the occurrence of synergism is generally facilitated in gemini
surfactant-conventional surfactant mixtures with respect to conventional surfactant-
conventional surfactant mixtures. Synergism in mixed micelle formation and surface
tension reduction efficiency and effectiveness was found in mixtures 1–4 in Table 1
(sulfonated gemini surfactant-tetradecyldimethylamine oxide mixtures) [12,13].
However, the same gemini surfactants showed no or little synergism with the nonionic
ethoxylated surfactant C12Ex owing to the protonation of the ether oxygen atoms of these
surfactants [13]. It was later shown that synergism could be obtained by reducing the
number x of ethylene oxide units in the surfactant head group [16]. Synergism in mixed-
micelle formation and surface tension reduction efficiency and effectiveness was also
found in mixtures of cationic gemini surfactants of the m-spacer-m type and alkyl
glucosides and maltosides (some of these systems are listed in Table 1, entries 23–25 and
27) [15]. No synergism was found when the gemini surfactant was replaced by the
Gemini surfactants 240
III.
PROPERTIES OF MICELLES IN MIXED GEMINI
SURFACTANT-CONVENTIONAL SURFACTANT SYSTEMS
A.
Aggregation Number and Ionization Degree of the Mixed Micelles
The aggregation numbers of mixed micelles of gemini and conventional surfactants have
been determined using small-angle neutron scattering (SANS) [25], time-resolved
fluorescence quenching (TRFQ) [17,20,26], and light scattering [19,27]. As is shown
here these results can be considered as preliminary in the sense that no real systematic
study has been reported as of today. Nevertheless, they reveal some useful trends.
The aggregation number of mixed micelles of the gemini surfactants 16-s-16 with s=3,
5 and 10 and of the conventional surfactant CTAB have been determined at a fixed
concentration of CTAB and increasing concentration of gemini surfactant [25]. CTAB
can be considered as the monomeric surfactant corresponding to the 16-s-16 surfactants
investigated. The results are represented in Fig. 4. The total number of hexadecyl chains
Mixed micellization between dimeric mixed micellization between dimeric surfactants 241
per micelle, NT, increases with the gemini surfactant concentration C (i.e., as the mixed
micelles become richer in 16-s-16 surfactant). At a given composition, NT increases in the
order 16–10–16<16–5–16<16–3–16, (i.e., in the order of increasing aggregation number
of the micelles of the pure gemini surfactant) [28,29].
A similar result was obtained in a study of the mixtures of the anionic gemini
surfactant Dim 1 (see chemical structure in footnote a in Table 1) with the nonionic
surfactants C12E5 and C12E8 [17]. The measurements were performed at a fixed total
surfactant concentration, much larger than the cmc of the mixture, in order that the
micelle composition remains very close to the weighing-in composition. Dim 1
progressively replaced the conventional nonionic surfactant. The results, represented in
Figs. 5 and 6, show large differences in the variations of NT with the gemini mole fraction
X for the two systems. The differences were attributed to the larger micelle aggregation
number and lower cloud temperature (about 32°C versus 77°C) of C12E5 with respect to
C12E8 and to the opposite variations with temperature of the aggregation number of Dim
1 and C12Ex surfactant micelles [17]. This explains
0.2–0.8 in the dimeric surfactant and to be smaller than the aggregation number of 12–2–
12 or C12E6 [19].
The reported results [17,25] suggested that the aggregation numbers in the mixed
systems are somehow a combination of the micelle aggregation numbers of the pure
surfactants. An attempt to interpret the data in Figs. 5 and 6 on the basis of ideal mixing
of the surfactants failed qualitatively and quantitatively [17].
Nonideal mixing was evident for the 12–2–12–C12E6 and 12–3–12–C12E8 mixtures
[20]. The variation of the micelle aggregation number with composition showed a
minimum, particularly for the 12–2–12–C12E6 mixture. However, this minimum was not
related to the occurrence of synergism in micelle formation [20]. The initial decrease of
the aggregation number of the nonionic surfactant caused by the addition of ionic
surfactant, up to a mole fraction of ionic surfactant of about of about 0.2, depends little
on the nature of the surfactant, whether conventional or dimeric. This decrease is due to
the increased electrostatic repulsive interactions between head groups upon introduction
of the ionic surfactant in the nonionic micelles.
A dynamic light-scattering study of 12–4–12 micellar solutions revealed the presence
of some large aggregates (size around 200 nm) coexisting with small spherical micelles.
The addition of DTAB resulted in the disappearance of the large aggregates, thus
revealing mixed-micelle formation [27].
Two studies dealt with mixtures of an anionic gemini surfactant and of an oppositely
charged conventional surfactant [26,30]. The first study is that of the mixture
[Na+,C12H25OP(O)(O−)O]2C6H4CH=CHC6H4/CTAB [30]. Only mixtures at low mole
fraction X of dimeric surfactant were investigated because precipitation occurred at
higher values of X. The authors postulated the formation of cross-linked CTAB-rich
micelles. The gemini surfactants would act as cross-links by having their two alkyl chains
embedded in the cores of two different CTAB-rich micelles. Large aggregates were also
detected by optical microscopy. The second study involved the very similar system
Dim1–CTAB [26]. The spacer group in Dim 1 (see footnote a in Table 1) is longer and
more flexible than in the gemini surfactant used in Ref. [30]. This should have favored
the formation of cross-linked micelles, should such micelles form. The TRFQ
investigation showed a very rapid and large increase of the micelle aggregation number
as the Dim 1 content of the mixed system was increased. Transmission electron
microscopy at cryogenic temperature (cryo-TEM) showed the presence of vesicles and of
very large and ill-defined structures at a Dim 1 mole fraction of 0.091, revealing that this
system was close to precipitation [26]. These large structures probably correspond to
those reported in Ref. [30]. Overall the results did not support the formation of cross-
linked micelles in mixtures of oppositely charged gemini and conventional surfactants at
low gemini surfactant content. Note that Monte Carlo simulations predicted the formation
of cross-linked micelles in mixtures of nonionic conventional surfactants and dimeric
surfactants [31]. It is likely, however, that the spacer must be extremely long for this
effect to occur.
The mixture of the cationic gemini surfactant 1,4-bis-(2′-(N-dodecylpyridino-4″-
yl)ethenyl)benzene dibromide with the oppositely charged SDS 7 could not be
investigated [32] Indeed, precipitation occurred readily and resulted in the formation of
microcrystals of a complex of the gemini surfactant dication and two dodecylsulfate ions.
Mixed micellization between dimeric mixed micellization between dimeric surfactants 245
Some information on the ionization of the mixed micelles can be found in the study of
the 16-s-16-CTAB mixtures by SANS [25]. The addition of 16-s-16 surfactants to CTAB
micelles was found to result in the formation of mixed micelles that are less ionized and
also larger than the pure CTAB micelles. The effect was more pronounced for the
addition of the gemini surfactant with the shortest spacer group.
B.
Shape of the Mixed Micelles
SANS studies showed that the spheroidal CTAB micelles are progressively transformed
into threadlike micelles upon addition of gemini surfactants 16-s-16. The surfactant with
the shortest spacer, s=3, was the most effective in achieving this transformation [25].
Conversely, cryo-TEM permitted the direct observation of the transformation of the
threadlike micelles formed by the 12–2–12 surfactant into spheroidal micelles upon the
addition of DTAB [33]. The transformation was completed at a mole fraction of DTAB
of about 0.3. Recall that DTAB can be considered as the monomer of 12–2–12 and that it
forms spherical micelles. It was later shown using SANS that the DTAB is distributed
almost uniformly in the mixed threadlike micelles, whereas it may have been assumed
that this spherical micelle-forming surfactant would be mostly localized at the
hemispherical end caps of the micelles [34].
Another study is that of the mixtures of Pluronic P105, an amphiphilic triblock
copolymer poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide)
(EO37PO56EO37), with the nonionic Surfynol 104 [35]. The chemical structure of Surfynol
104 is [(CH3)2CHCH2C(CH3)OH]2C≡C, showing that it is a dimeric diol surfactant with
a rigid C≡C spacer. The addition of Surfynol 104 to a micellar solution of Pluronic P105
was found to induce dramatic variations of rheological properties that reflected a
transformation of the initially spheroidal micelles of pure Pluronic P105 into mixed
threadlike micelles. Further additions of Surfynol 104 resulted in the formation of a
lamellar structure. The formation of threadlike micelles gave rise to an increase of the
solution viscosity by a factor 104, as can be seen in Fig. 7 [35]. The viscosity is seen to be
slightly affected by the addition of Surfynol 104 up to a weight ratio of about 0.3; then, it
increases dramatically as the mixed micelles turn threadlike. The second increase of
viscosity at a high weight ratio corresponds to the formation of the lamellar structure.
This behavior is very interesting because Surfynol 104 is only sparingly soluble in water.
However, it is very soluble in the presence of Pluronic P105, giving rise to solutions of
tunable viscosity that are of interest for many applications.
In Eq. (4), ηGem and ηCTAB are the microviscosities of the pure gemini and conventional
surfactant micelles, respectively. However, we checked that Eq. (4) does not account for
the results and is thus not valid.
Gemini surfactants 246
C.
Microviscosity of the Mixed Micelles
The fluorescence anisotropy of the fluorescent probe 1,6-diphenylhexatriene has been
used to investigate the microviscosity of mixed micelles in the 16-s-16-CTAB systems
[25]. The addition of 16-s-16 surfactants with s=3, 5, and 10 to a micellar solution of
CTAB was found to result in a linear increase of the micelle microviscosity with the
mixture composition up to a mole fraction of gemini X=0.33. The increase of
microviscosity was the largest with the gemini surfactant forming the micelles having the
largest microviscosity [25,28]. This behavior suggests that the microviscosity of the
investigated mixtures may obey an ideal mixing law of the type
(4)
Mixed micellization between dimeric mixed micellization between dimeric surfactants 247
IV.
MICROSTRUCTURE OF THE MIXED SOLUTIONS OF GEMINI
AND CONVENTIONAL SURFACTANTS
Cryo-TEM [8] has been the technique used for all studies of microstructure of mixtures
of gemini and conventional surfactants. The first study involved the qualitative
observation of the change of microstructure of 12–2–12 solutions
V.
CONCLUSIONS
The reported studies of mixtures of gemini and conventional surfactants have revealed
that synergism in mixed micellization and in surface tension reduction efficiency and
effectiveness is more easily obtained than in mixtures of conventional surfactants. This
result is a very important one for future uses of gemini surfactants in formulations. The
aggregation behavior of mixtures of gemini and conventional surfactants and the
microstructure of the solutions of these mixtures have started to be investigated. The
reported results show no unusual behaviors with respects to conventional surfactant
mixtures at this stage. Nevertheless, additional studies are required to further understand
the conditions that determine synergism. Also, to the best of our knowledge, there has
been no report on the solubilization by mixtures of gemini and conventional surfactants.
Such studies, although tedious, remain necessary for a full evaluation of gemini
surfactants in future uses.
REFERENCES
1. Holland, P.M., Rubingh, D.N., Eds.; Mixed Surfactant Systems, ACS Symposium Series No.
501; Washington DC: American Chemical Society, 1991.
Gemini surfactants 250
2. Ogino, K., Abe, M., Eds.; Mixed Surfactant Systems; New York: Marcel Dekker, 1993.
3. Kwetkat, K. J. Cosmet. Sci. 2001, 52, 414.
4. Kwetkat, K. SOFT (Eng. Ed.) 2002, 128, 38.
5. Zana, R. In Surfactant Solutions. New Methods of Investigation; Zana, R. Ed.; New York:
Marcel Dekker, 1987. Chap. 5.
6. Cabane, B. In Surfactant Solutions. New Methods of Investigation; Zana, R. Ed.; New York:
Marcel Dekker, 1987. Chap. 2.
7. Candau, S.J. In Surfactant Solutions. New Methods of Investigation; Zana, R. Ed.; New York:
Marcel Dekker, 1987. Chap. 3.
8. Talmon, Y. In Modern Characterization Methods of Surfactant Systems; Binks, B. Ed.; New
York: Marcel Dekker, 1999; p. 148.
9. Rosen, M.J.; Zhu, B.Y. J. Colloid Interface Sci. 1986, 99, 427.
10. Zhu, B.Y.; Rosen, M.J. J. Colloid Interface Sci. 1986, 99, 435.
11. Rosen, M.J.; Murphy, D.S. J. Colloid Interface Sci. 1986,110, 224.
12. Rosen, M.J.; Zhu, Z.H.; Gao, T. J. Colloid Interface Sci. 1993, 157, 254.
13. Rosen, M.J.; Gao, T.; Nakatsuji, Y.; Masuyama, A. Colloids Surfaces A 1994,88, 1 .
14. Liu, L.; Rosen, M.J. J. Colloid Interface Sci. 1996, 179, 454.
15. Li, F.; Rosen, M.J.; Sulthana, S.B. Langmuir 2001, 17, 1037.
16. Rosen, M.J.; Zhou, Q. Langmuir 2001, 17, 3532.
17. Zana, R.; Levy, H.; Kwetkat, K. J. Colloid Interface Sci. 1998, 197, 370.
18. Zhao, J.; Christian, S.D.; Fung, B.M. J. Phys. Chem. B 1998, 102, 7613.
19. Esumi, K.; Miyazaki, M.; Arai, T.; Koide, Y. Colloids Surfaces A 1998, 135, 117.
20. Alargova, R.G.; Kochijashky, I.I.; Sierra, M.L.; Kwetkat, K.; Zana, R. J. Colloid Interface Sci.
2001, 235, 119.
21. Holland, P.M. In Mixed Surfactant Systems; Holland, P.M., Rubingh, D.N., Eds.; ACS
Symposium Series No. 501. Washington DC: American Chemical Society, 1991; p. 40.
22. Oda, R.; Huc, Y.; Danino, D.; Talmon, Y. Langmuir 2000, 16, 9759.
23. Rosen, M.J.; Tracy, D.J. J. Surfact. Detergents 1998, 1, 547.
24. Funasaki, N. In Mixed Surfactant Systems’, Ogino, K., Abe, M., Eds.; Marcel Dekker: New
York, 1993; p. 145.
25. De, S.; Aswal, V.K.; Goyal, P.S.; Bhattacharya, S. J. Phys. Chem. B 1997, 101, 5639.
26. Zana, R.; Levy, H.; Danino, D.; Talmon, Y.; Kwetkat, K. Langmuir 1997, 13, 402.
27. Pisarcik, M.; Dubnickova, M.; Devinsky, F.; Lacko, I.; Skvarla, J. Colloids Surfaces A 1998,
143, 69.
28. De, S.; Aswal, V.K.; Goyal, P.S.; Bhattacharya, S. J. Phys. Chem 1996, 100, 11664.
29. Aswal, V.K.; De, S.; Goyal, P.S.; Bhattacharya, S.; Heenan, R.K. Phys. Rev. E 1998, 57, 776.
30. Menger, F.M.; Eliseev, A.V. Langmuir 1995, 11, 1855.
31. Maiti, P.K.; Kremer, K.; Flimm, O.; Chowdhury, D.; Stauffer, D. Langmuir 2000, 16, 3784.
32. Stathatos, E.; Lianos, P.; Rokotoaly, R.H.; Laschewsky, A.; Zana, R. J. Colloid Interface Sci.
2000, 227, 476.
33. Talmon, Y.; Zana, R. Nature 1993, 362, 228.
34. Schosseler, F.; Anthony, A.; Beinert, G.; Zana, R. Langmuir 1995, 11, 3347.
35. Guo, L.; Colby, R.H.; Lin, M.Y.; Dado, G.P. J. Rheol. 2001, 45, 1223.
36. Alexandridis, P.; Hatton, A.T. Colloids Surfaces A 1995, 96, 1.
37. Danino, D.; Talmon, Y.; Zana, R. J. Colloid Interface Sci. 1997, 185, 84.
38. Bernheim-Groswasser, A.; Zana, R.; Talmon, Y. J. Phys. Chem. B 2000, 104, 12192.
11
Special Gemini Surfactants: Nonionic,
Zwitterionic, Fluorinated, and Amino Acid
Based
TIM W.DAVEY Dulux, Melbourne, Victoria, Australia
I.
INTRODUCTION
Most solution studies of gemini surfactants have focused on cationic geminis largely
because of their stability and ease of synthesis. For example, the m-s-m type cationic
gemini surfactants [alkanediyl-α,ω-bis(alkyldimethylammonium bromide) gemini
surfactants, m and s are the carbon numbers of the alkyl and alkanediyl groups], which
are prepared easily in one reaction step, have been widely studied and this work has
helped establish the basic principles of gemini surfactant behavior. The majority of work
on gemini surfactants has been carried out over the past decade, and the area has “come
of age” in recent years, with a huge increase in the range of surfactants and end uses
being studied. The collaboration of organic chemists with other disciplines has led to the
preparation and examination of a wide variety of molecular architectures. Gemini
surfactants are moving from laboratory curiosities to alternatives to conventional
surfactants in many fields of study. Whereas such surfactants were once prepared and
simply examined in dilute solutions, they are now found as the subject of many different
fields, such as in gene transfection and antimicrobial studies, among many other diverse
areas (see Chapters 12 and 13).
This chapter explores lesser known types of gemini surfactant: nonionic, zwitterionic,
fluorinated, and amino acid-based surfactants. In many cases, these surfactants require
multistep syntheses, and a quick perusal of the chemical structures provided in this
chapter demonstrates that future work is limited only by the surfactant chemists’
imagination. This wide variety of chemical structure is the main focus of this chapter.
Although the bulk of work covered here is of an academic nature, a number of patents
have been filed and these are also discussed. Several gemini surfactants are now
commercially available, although the added expense associated with the preparation of
many of these compounds (compared to conventional surfactants) will probably lead to
their use only in specialist applications.
Nonionic gemini surfactants contain a hydrophilic component based on either
poly(ethylene oxide) (PEO) or a carbohydrate unit. There have been few academic
studies on PEO-based gemini surfactants, probably due to the difficulties associated with
handling ethylene oxide during their preparation. The industrial use of ethylene oxide is
routine and, therefore, PEO-based geminis represent a convenient starting point for the
Gemini surfactants 252
II.
POLY(ETHYLENE OXIDE)-TYPE GEMINI SURFACTANTS
A.
Papers
The most common type of conventional nonionic surfactant in use is of the
alkylpoly(ethylene oxide) type, in which the alkyl unit may also include an aromatic
moiety, such as in the case of the well-known alkylphenylethoxylates (APEs).
Gemini surfactant (1) was prepared in three steps [2]. Surfactant (1a) was found to be
insoluble in water, whereas (1c) was deemed not pure enough, and therefore neither were
investigated further. Comparative studies of (1b) with its monomeric equivalent were
conducted. A plot of cloud temperature versus concentration of (1b) gave a minimum of
48.5°C at 0.5 wt% surfactant, whereas the monomer showed no clouding up to 85°C.
Thus, the “dimerization” of the monomer causes a large decrease in the cloud
temperature (of at least 35°C), perhaps because in the case of the dimer, two moieties are
simultaneously involved in intermolecular interactions leading to clouding, as opposed to
one in the case of the monomer.
Special gemini surfactants 253
Scheme 1–7
The critical micellization concentration (cmc) of (1b) was 0.02 mM, whereas the cmc of
the monomer was 0.4 mM, or 20 times higher than the gemini (or 10 on a per-chain
basis), as is found for ionic geminis and their monomers [3]. At low temperature (5°C),
the two surfactants have, within error, the same value of N (aggregation number), but for
(1b), it was found that N became larger more rapidly with increasing temperature than the
monomer. This was reasoned to be due to the lower cloud point of (1b), which led to
micelle growth at a lower temperature.
FitzGerald et al. (unpublished results) prepared and studied the gemini surfactant (2).
The short (x=5) poly(ethylene oxide) derivatives were found to form a liquid isotropic
phase coexisting with a lamellar phase at room temperature and were not studied further.
The cmcs of surfactants (2) (x>5) ranged from 1.5×10−7 to 6.5×10−7 M, as determined by
Gemini surfactants 254
pyrene fluorescence (I1/I3). The extremely low values of the cmc make them somewhat
unreliable, but they were deemed of the correct order of magnitude based on
Scheme 8–12
an extrapolation of the cmcs from much higher concentrations in a mixed-surfactant
system using ideal-mixing theory. Attempts at measuring the cmc’s by surface
tensiometry were hampered by problems arising from significant adsorption on the walls
of the vessel containing the surfactant solution, and the results were unreliable. Small-
Special gemini surfactants 255
angle neutron scattering (SANS) measurements showed that 1 wt% solutions of the
surfactants with larger head groups (x≥15) generally formed spherical micelles, whereas
those with smaller head groups (x=10) formed elongated micelles.
A number of nonionic gemini surfactants have been prepared by Fukunaga et al. [4–6]
and complexed with a lipase enzyme to study the catalysis of transesterification reactions
in organic solvents. Surfactants prepared and studied in this work include the PEO-based
(3) and (4) [6], as well as carbohydrate-based compounds [4,5] (see Section III).
B.
Patents
A number of patents have been granted to Rhodia for PEO-based nonionic gemini
surfactants [7–13]. Most of these claim the geminis as extremely effective emulsifiers for
oil-in-water emulsions at very low surfactant concentrations. Illustrative examples
include the polyaromatic (5) [10] which has a cmc of 7.9×10−7 M, and (6) [8] with a cmc
of 1.9×10−5 M.
Nonionic gemini surfactants were claimed for use in cosmetics with low skin irritancy
and good storage stability. An example provided was compound (7) [14].
Air Products and Chemicals Inc. has patented the general acetylenic diol structures (8)
[15] and (9) [16]. Surfactant (8) is claimed for use in waterborne coatings, whereas (9) is
for use in a range of water-based compositions, including coatings and inks.
III.
CARBOHYDRATE-BASED GEMINI SURFACTANTS
A number of gemini surfactants based on carbohydrates have been prepared and studied.
The vast majority of these are derivatives of β-D-glucose and these will be discussed
first, followed by other types.
A.
Papers
Gemini surfactants based on alkyl glucopyranoside, such as compounds (10), were
prepared via the reaction of suitably protected alkyl glucopyranosides with the
corresponding diacyl dichlorides [17–22].
The cmc of the butyl derivatives (m=4) were 6 mM [X=(CH2)2] and 8 mM [X=(CH2)3]
compared to 75 mM for the monomeric surfactant butyl-α-D-glucopyranoside. The butyl
surfactants linked through O-2 were also pre pared [18] and shown [21] to have cmc’s
approximately threefold to fourfold lower than the O-6-linked surfactants (10). The cmc
of (10) with m=4 and X=(CH2)2 was 1.5 times lower than the corresponding β-anomer
and a similar ratio was found for the monomeric surfactants, showing that the anomeric
configuration has an effect on the aggregation behavior of these types of surfactant [20].
The authors suggest that the β-anomer is able to better pack inside the micelle [21]. It was
found that the cmc’s of all compounds in an homologous series [X=(CH2)2, m=4, 8, 12,
14] were all quite similar (1.8–7.2 mM), contrary to that found for conventional
Gemini surfactants 256
surfactants with varying hydrophobic chain lengths. Possible explanations for these
results include self-coiling or premicellar aggregation [21].
Compounds of type (11) were prepared [23–25] in which the head group and
hydrocarbon tail are linked through a single carbon atom. The solution behavior of
compounds with n=4 and m=5–9 was investigated. The phase diagrams exhibited
behavior typical of conventional nonionic surfactants, except for the compound with m=7
in which an upper critical solution temperature (UCST) was observed from 0.4 to 8 wt%.
This may constitute the first observation of an UCST for a nonionic surfactant. The
Krafft temperatures for n=5 and 6 were below 5°C and for n=1 and 8 were 34°C, whereas
the compound with n=9 was insoluble up to 100°C. The cmc’s, determined by surface
tensiometry, varied from 1.05×10−2 M (for n=5) to 4.2×10−5 M (for n=8) and did not
change appreciably with temperature. The areas per molecule at the air-water interface
were 0.68–0.81 nm2. SANS studies of the micellar phases revealed that cylindrical
micelles are present for n=5–7 and discoidal micelles for n=8 [25]. These compounds
were also the subject of an earlier patent by Briggs and Pitt (see Section III.B for further
details) [26].
Surfactant (12), similar in structure to (11), was synthesized using pentaerythritol as
the linker between the hydrophobic alkyl chains and functionalized β-D-glucose [27].
The surface pressure-area isotherm measured at the air-water interface showed two
distinct regions in which the surfactant is either in an expanded or condensed state. The
discontinuity between the two states was found to occur at higher pressures as the
temperature was increased.
Gemini compounds (13) were prepared in two steps from β-D-glucose [28, 29]. The
Krafft temperatures generally lie in the range 30–40 °C, and their solution properties
were examined by the addition of 5 mol% of sodium dodecanesulfonate (to lower the
effective Krafft temperature). Compounds with long spacer chains (e.g., m=13, s=10)
generally formed vesicles, whereas shorter spacers gave rise to threadlike micelles and
other structures, as observed by electron microscopy. A 5-mM solution of compound (13)
(m= 13, s=6) was observed to be clear at 55°C but turned blue after 30 min at 60°C.
When cooled to room temperature, the solution was clear, but had
Special gemini surfactants 257
Scheme 13–17
become viscoelastic, which the authors attributed to the formation of strongly entangled
threadlike micelles.
Amine-based surfactants (14) were synthesized in two steps by the reaction of β-D-
glucose with a diamine, followed by alkylation with an aldehyde [30– 32]. Such
surfactants are protonated at low pH (and therefore cationic) and form micelles in
aqueous solution, whereas at a pH greater than approximately 5.5, the surfactants are
nonionic and forms vesicles in solution, upon sonication. The dependence of morphology
on pH was investigated as a means of delivery of DNA to eukaryotic cells. The proposed
method involved the DNA being enclosed in vesicles at physiological pH, and upon
transport through the cell membrane (pH of 2–4), the vesicles would undergo a transition
to micelles, thereby releasing their contents. Transfection efficiencies in vitro using
Gemini surfactants 258
surfactant (14) (with an oleyl tail) were up to 2.7-fold greater than for a standard
commercial product. The association behavior was also simulated by molecular dynamics
and self-consistent field calculations [32]. It was shown that the titratable head group was
responsible for the micelle to vesicle transition and that the aggregate morphology was
very sensitive to ionic strength. These compounds were also the subject of an earlier
patent by this group [33].
A more detailed examination of the pH-dependent solution aggregation behavior of
compound (14) (m=16, s=6) was carried out [30]. It was shown that a transition from
micelles to vesicles occurs with increasing pH, with an intermediate region of pH of
approximately 4–5.5. Plots of both surface tension and turbidity against pH showed a
change in the intermediate region. The solution appeared clear at low pH, bluish above
pH 5.5, and milky at pH 7.5. The area per head group at pH 6 was 0.69 nm2 , as
determined by surface pressure versus area measurements, as compared to 1.09 nm2
determined by surface tension measurements. This change is consistent with increased
repulsion between head groups for the cationic surfactant. Vesicles up to 2 µm in size
were visualized by optical microscopy and cryo-TEM (transmission electron microscopy)
at pH 7.5.
A series of glucose- and lactose-based surfactants (15) have been prepared and
complexed with lipase enzymes to study the catalysis of transesterification reactions in
organic solvents [4,5,34]. The hydrophile-lipophile balance (HLB) values of the gemini
surfactants were calculated and it was found that those with the lowest HLB values
(R=H, m=16 and 18) were insoluble, whereas all others were easily soluble. The gemini
surfactants always showed higher enzymatic activities (with four different enzymes) than
two-tailed or two-headed surfactants of the same HLB value. The D-isomer of (15)
(different stereochemistry at the asterisk) showed much less enhancement of activity than
the L-form with the enzyme Candida rugosa lipase. This difference was related to the
conformational fit of the surfactants with the enzyme [34].
A number of surfactants (16) based on β-D-galactose were prepared in seven steps
[35]. These compounds contain four galactose units in the head group with hydrocarbon
or fluorocarbon alkyl chains as the hydrophobic moiety. Two surfactants (not shown)
were also prepared in which the alkyl spacer was replaced with an aromatic unit, but
these were found to be poorly soluble.
The cmc’s were generally found to be quite low. With R=C11H23, the cmc’s were
about 60 times lower than for a monomeric equivalent. The surface tension of solutions
of the gemini surfactants took a long time (0.5–1 hr) to reach an equilibrium value. This
equilibration could be accelerated by increasing the temperature to 50 °C, but the cmc
surprisingly increased by a factor of over 100 at this temperature. In contrast to
conventional surfactants [36], it was found that an increase in the length of the
hydrocarbon tail led to an increase in the cmc [e.g., 7.5×10−5 M (R=C17H35, s=4)
compared to 7.5×10−7 M (R=C11H23, s=4]. A similar time dependence of the surface
tension and unusual cmc effects with the variation of the alkyl chain length have been
observed previously for other gemini surfactants and attributed to premicellar aggregation
(see Chapter 5). Indeed, plots of surface tension versus concentration for R=C17H35 show
two breaks, the first being attributed to premicellar aggregation.
The cmc’s of surfactants (16) with a fluorocarbon tail were in the range (2.8–5)×10−5
M, somewhat higher than their hydrocarbon analogs with R=C11H23. This result is
Special gemini surfactants 259
surprising, as the C2H4C8F17 group would be expected to resemble (for cmc comparison
purposes [37]) a C14H29 group, given the greater hydrophobicity of fluorocarbon groups,
and therefore give rise to lower cmc values. Finally, sonication of solutions of geminis
(16), in most cases, produced vesicular and fibrous aggregates as seen by electron
microscopy.
Surfactants of type 17 were prepared in four steps [38]. These compounds were
generally nontoxic to Gram-negative bacteria and fungi, but inhibited the growth of some
Gram-positive bacteria. They could be biodegraded by micro-organisms, although could
not be termed “readily biodegradable” according to OECD standards. The octyl
derivatives exhibited stronger antimicrobial activity and, as a result, slower
biodegradation compared to the dodecyl compounds. Of the different head groups, the
glucoheptonyl compounds degraded the slowest. All compounds were markedly slower
to biodegrade than the monomeric analogs N-dodecyl-N-methylaldonamides [38]. Small-
angle neutron scattering (SANS) studies on (17) (X=lactobionyl, m= 8) were used to
measure the size and shape of the solution micelles and their variation with concentration
and temperature [39].
Surfactants (18) based on trehalose (a disaccharide of glucose) have been prepared in
four steps from trehalose [40]. This series exhibited low water solubility, which the
authors suggest may be due to intramolecular and intermolecular bonding within the
crystal. Pressure-area isotherms for monomolecular films of m=11 and 17 at the air-water
interface gave areas per molecule of 1.20 nm2 for both compounds, as compared to about
0.49 nm2 for alkyl glucosides [41]. The similarity in the isotherms for the surfactants with
m=11 and 17 led the authors to the conclusion that the
Gemini surfactants 260
Scheme 18–23
trehalose spacer dominates the interfacial packing. Vesicles were formed in water at
room temperature by some of these compounds [m=9, 11, 17 (cis-9)]. The others
compounds in the series (m=13, 15, 17) had transition temperatures above 33 °C.
Molecular modeling demonstrated that surfactants (18) adopt a tubular shape, which
explains their tendency to prefer bilayer packing.
Rose et al. [42] prepared 15 gemini surfactants based on trehalose structure (19),
where R1 and R2 are a range of alkyl substituents, including alkylamino and sulfonamido
derivatives. Significant antimycobacterial activity was reported against Mycobacterium
tuberculosis H37Ra and a panel of clinical isolates of M.avium.
Special gemini surfactants 261
B.
Patents
Compounds (21a–21b) were prepared, beginning with lactose and spermine, and claimed
as useful dispersing agents for crystalline drugs [47]. Compounds (21a–21b), as well as
other variants, were also prepared by another group which they claimed for use in
gelatin-containing photographic emulsions [48].
Also claimed for photographic materials were compounds of type (11) [26]. In an
example given, the compound with m=6 and n=3 had a cmc of 0.08 wt % (1.4 mM)
compared to 3 wt% (0.1 M) for a monomeric equivalent (2-ethylhexylgluconamide), a
difference of greater than 70-fold.
Compounds of type (14) were claimed as gene transfer agents, as discussed earlier
[33].
A range of carbohydrate-based surfactants, such as (22), were claimed as suitable for
use as detergents [49–51]. Tetrameric variants of these surfactants were also the subject
of a patent by the same authors [52].
Tracy et al. [53] have claimed a range of nonionic gemini surfactants based on using
nitrogen as a linker between the spacer, hydrophobe, and hydrophile, such as
lactosamine-derived (23), which has a cmc of 1.7×10−6 M.
Also prepared by Tracy et al. [54] were a series of surfactants in which the linker atom
is carbon which is connected to the hydrophilic portion via a carbonyl group. An example
provided was that of N-methylglucamine compound (24), for which a cmc of 1.2×10−5 M
was measured.
A Japanese patent claims trehalose surfactants similar in structure to (18) [55] as
antitumor agents.
A patent was also obtained for compounds of type (17) [56].
Gemini surfactants 262
Scheme 24–28
Special gemini surfactants 263
IV.
ZWITTERIONIC GEMINI SURFACTANTS
A.
Papers
The zwitterionic surfactants of type (25), containing a phosphodiester anion and
quaternary ammonium cation, were synthesized in two steps [57–62].
These surfactants form a variety of structures in solution, depending on the lengths of
the alkyl chains. A “structural phase diagram” was constructed [59] relating the
spontaneous aggregate structure at low concentration (less than 5 wt%) to the values of m
and m′ for 42 surfactants. For surfactants with alkyl chains of similar length, micelles
were observed for short chains (values of m+m′ less than approximately 18) and
polydisperse vesicles for long chains (m+m′ greater than approximately 20). These
vesicles ranged in size from 30 nm up to 100 µm. For intermediate values of m+m′, a
spongelike coacervate was present, whereas gels were formed when the alkyl chains were
quite different in length (m and m′ differed by greater than approximately 8). The gels
were formed from an interconnected network of vesicle-sized particles.
The coacervate phases, formed in equilibrium with a dilute surfactant phase, were 82–
86% water, with the sponge framework occupying the entire volume of the phase. Other
systems have been depicted as spongelike, although they have never been imaged as
such, whereas for this system, high-resolution scanning electron microscopy images show
true spongelike structure. This is also the first case of a binary sponge-forming system, as
compared to ternary or quaternary systems [58].
The cmc’s of these compounds were measured by surface tensiometry and
fluorescence. The cmc for the surfactant with m=10 and m′=12 was 20 times lower than
for the catanionic system comprising a 1:1 mix of dodecyltrimethylammonium bromide
(DTAB) and sodium decylphosphate (SDP). The kinetics of organization at the air-water
interface were quite slow, and up to 2 h were required for the surface tension of solutions
of some of the surfactants to reach an equilibrium value. A more detailed study was
conducted on four surfactants of type (25), with m and m′ equal to 14 and 8, or 12 and 10,
or vice versa (i.e., m+m′=22, m not equal to m′)[57,60–62]. These combinations represent
symmetrical or asymmetrical systems. The cmc’s of these surfactants are approximately
10−5 M and approximately double upon addition of an equimolar amount of NaBr. The
surface area per molecule at the air-water interface calculated from the surface tension
data (using a value of n=1 in the Gibbs equation) were approximately 0.30 nm2, much
lower than for the DTAB-SDP catanionic system, or DTAB or SDP alone. This suggests
that the surface monolayer formed by (25) is very densely packed. The addition of NaCl
gave even lower surface area values.
The adsorption of (25) at solid-solution interfaces was studied by reflectometry, using
either a hydrophilic surface (silica) or a hydrophobic surface (silica that had been treated
with dichlorodimethylsilane). The amount of surfactant that adsorbed on the hydrophilic
surface was about double than on the hydrophobic surface, suggesting that micelles are
present on the former and a monolayer or hemimicelles on the latter. For both surfaces,
Gemini surfactants 264
an increase in the surfactant concentration above the cmc resulted in an increase in the
amount of adsorbed material, implying some kind of aggregation was occurring on the
surface. The amounts further increased with the addition of NaCl. On the hydrophobic
surface, at a concentration in the vicinity of the cmc, the asymmetrical surfactants had
much higher surface areas per molecule than the symmetrical surfactants [62].
Examination of the surfactant (25) with m=14 and m′=8 by pulsed-gradient spin-echo
nuclear magnetic resonance (NMR) showed that the rate of monomer exchange between
aggregate and solution was slower than the NMR time scale (approximately 100 ms) and
that the aggregate had a mean size of 55 nm [60].
The ability of the geminis (25) to stabilize polystyrene latex was investigated, and it
was found that the geminis provided better heat and salt tolerance compared to
conventional surfactant systems of anionic and non-ionic surfactants (or mixtures thereof)
[61].
Compounds of type (26) were synthesized in three steps [63]. Upon sonication in
solution, they self-assemble into mainly small unilamellar vesicles. Hydrolysis of (26)
occurred at pH~3 to give a cationic surfactant (which forms micelles) and a keto acid
(which precipitates). Thus, vesicles of surfactant (26) have potential as entrapment and
release devices: A compound could be included within the vesicle interior, which would
then be released when the vesicle entered an acidic environment and broke up.
A betaine-based surfactant (27) was prepared with a dimer of linoleic acid as the
hydrophobic portion. The cmc is 0.1 mM at 25 °C. Viscoelastic behavior is observed at
low surfactant concentrations and is indicative of wormlike micelles. The linear flow
properties of (27) were examined and it was found that the rheological properties (such as
zero shear viscosity and relaxation times) are mainly controlled by diffusion (reptation)
processes, as compared to kinetic (breaking) processes. This is in contrast to viscoelastic
anionic and cationic systems in which kinetic processes dominate. The results suggest
that the micellar aggregates are linear and not branched. The addition of increasing
amounts of NaCl leads to gel forming and ultimately pure reptation behavior [64–66].
B.
Patents
A small number of betaine-derived gemini surfactants have been patented. Compounds
with the generalized structure (28), prepared in two steps, are claimed for use as
emulsifiers, antifoaming agents and detergents [67].
Special gemini surfactants 265
Scheme 29–39b
Compounds such as (29) were synthesized in three steps and claimed for use in a
range of products such as shampoos and detergents, having good skin feel and reduced
skin and hair irritation [68].
Also claimed for use in shampoos are compounds including (30), which is also stated
to have bactericidal activity and pronounced eye compatibility. Compound (30) was
prepared in three steps from triethylenetetraamine [69].
Gemini surfactants 266
V.
FLUORINATED GEMINI SURFACTANTS
Two m-2-m-type surfactants were prepared in which either one (31) or both (32) of the
alkyl chains contained a fluorocarbon portion. These surfactants were prepared in five
steps, using iodoperfluorooctane as the source of the fluorocarbon tail [70].
The solution aggregation behavior of these surfactants was examined, both alone and
in mixtures with each other or the hydrocarbon gemini surfactant 12–2–12. The cmc’s of
(31) (2×10−4 M) and (32) (28×10−6 M) are comparable to the cmc’s of hydrocarbon
gemini surfactants 16–2–12 and 16–2–16, respectively, as would be expected from the
equivalence (for cmc purposes) of 1 CF2 to 1.5 CH2 units [37]. 1H- and 19F-NMR showed
two signals for each system above the cmc, indicating slow exchange (compared to the
NMR timescale) of monomeric surfactant between the aggregates and the bulk solution.
Cryo-TEM revealed the presence of vesicles in both systems above the cmc, along with
entangled micelles for (31) in the range 2–5 mM, resulting in a solution that was
viscoelastic.
When the hybrid surfactant (31) was mixed on an equimolar basis with either (32) or
12–2–12, one cmc was measured by conductivity. Fitting of the cmc data using regular
solution theory gave values of the interaction parameter (β) of −0.84 and −1.24,
respectively, indicating nonideal mixing. When (32) was mixed with 12–2–12, two
breaks in the plot of conductivity versus concentration were observed, consistent with the
formation of fluorocarbon-rich and hydrocarbon-rich micelles.
Other fluorocarbon-containing gemini surfactants that have been prepared are cationic
compounds (33–36) and anionic (37).
No physico-chemical properties were reported for (33) [71], whereas (34) [72] and
(35) [73] formed vesicular aggregates upon sonication. The cmc values for compounds of
type (36) were in all cases approximately 0.1 M. This apparent invariance of the cmc with
fluorocarbon chain length, spacer length, and heteroatom type was attributed to
premicellization effects [74]. The anionic surfactants (37) were prepared in two steps and
the surface tensions of 0.5-g/L solutions were 25 mN/m, except for the compound with
m=8, which was insoluble [75].
Cationic gemini surfactants with structures akin to (33–36) have also been investigated
for biocide activity against Gram-positive and Gram-negative bacteria, and although
some showed promising activity, little detail was given [76].
The surfactant (38) contains a fluorocarbon portion within the spacer group and was
prepared in three steps [77].
Fluorocarbon surfactants have also been prepared containing β-D-galactose head
groups [see compound (16) in Section III].
VI.
AMINO-ACID-BASED GEMINI SURFACTANTS
A large body of work has been carried out on the arginine-derived surfactants (39a–39b).
These compounds were synthesized in three steps, beginning with nitroarginine and an
alkanoyl chloride [78,79]. The preparation of (39) has also been carried out
Special gemini surfactants 267
exhibit a broad range of antimicrobial activity and are generally more effective than the
monomeric surfactants [79,88]. The aquatic toxicity is lower than for conventional
quaternary ammonium surfactants, whereas the surfactants (39a-39b) with s≤6 are
readily biodegradable [88].
Compounds (39b) (s=6, 9) were investigated as sequestering agents of
lipopolysaccharides (LPSs), an endotoxin on the outer membrane of Gram-negative
bacteria, which induces septic shock syndrome. LPS contains a glycolipid portion (lipid
A) which includes bisphosphate amphiphilic segments that may bind cationic surfactants,
thereby preventing septic shock. Compound (39b) was found to have similar sequestering
abilities to a known effective inhibitor (an amphiphilic derivative of spermine). This
result was also confirmed by studies in mice, but further work is needed to reduce the
cytotoxicity of these gemini surfactants [89].
The peptide-based cationic gemini surfactants (41–44) were prepared primarily to
investigate their behavior as gene-transfer vectors for delivery of DNA into mammalian
cells. Conventional cationic surfactants, such as cetyltrimethylammonium bromide, are
known to complex with DNA for in
Special gemini surfactants 269
Scheme 40–44
vitro delivery of genes to cells. The gene transfection efficiency of the gemini surfactants
was typically determined by their ability to mediate the transfer in vitro of a luciferase
reporter gene across Chinese hamster ovary cell membranes [90–96]. Such compounds
are also claimed in patents for use as transfection agents [97,98].
Gemini surfactants 270
Eighteen compounds of type (41) were prepared, with variations in the alkyl chain
length and nature of the cationic peptide head group [90– 92,95,97]. The backbone of the
surfactant was formed from the amino acids serine and cysteine. The peptide unit
generally consisted of three lysine residues linked either in series via the α- or ε-amino
group or with two lysine residues linked to both the α- and ε-amino groups of the first.
Surfactants with one and two lysine residues were also prepared. The peptide moiety thus
had between two and four amino groups which could be protonated. The cmc of
compounds (41) with a dodecyl hydrophobic chain was measured as 0.3 ± 0.1 mM,
although no method was stated [91].
Best gene transfection efficiency was found with ε-linkages compared to α-linkages,
whereas increasing the number of lysine residues (corresponding to more cationic
ammonium groups) generally led to improved activity. The authors suggest that the
DNA-surfactant interaction must be at least partially electrostatic and depends on the
spacing of the ammonium groups. An increase in the hydrophobic chain length also gave
rise to improved activity. Comparable or better levels of activity to a commercial product
were observed when the gemini surfactants were combined with a basic polypeptide.
Single-chain analogs based on S-methylcysteine were inactive. The transfection ability of
compounds (41) was related to the structure of the aggregate they formed on mica in
dilute solution (1 mg/mL). Those with low transfection efficiency formed fibrils, whereas
those with higher activity formed irregular-shaped aggregates made up of smaller
structures [99].
Surfactants based on tartaric acid were prepared, in which the head group contains the
groups ethylenediamine (42a) or lysine (42b), or both (42c). These surfactants were
shown to mediate the transfection of the luciferase gene but were also considerably
cytotoxic. Circular dichroism studies showed some interaction with DNA for (42b) and
(42c), although the secondary structure of the DNA was barely affected. Binding with
DNA was detected by the ability of the DNA to release ethidium bromide when
combined with all three gemini surfactants. The molecular area occupied at the air-water
interface increases in the order (42a)<(42b)<(42c). Platelike structures were observed in
solution upon sonication of (42a) and (42b), whereas no aggregates were observed for
(42c) [94].
Surfactants based on the tetraamine spermine were prepared in which the hydrophobic
group is attached to either the middle (43a) or terminal nitrogen atoms (43b–43c), or
linked to the terminal nitrogen atoms via a lysine residue (structure not shown) [93,96].
In the case of the oleyl derivatives (43a–43b), greater transfection activity was
observed for (43b) in which the oleyl groups are attached to the terminal nitrogens. The
head-group moiety X was composed of either a single lysine, or one serine and three
lysine residues. Transfection activity was measured for four cell lines, and surfactants
with the lysine head group had the greatest activity against three of these cell lines. This
result demonstrates that different structural features are optimal for different cell types,
and as opposed to the results for (41), increasing the number of cationic groups does not
always provide greater transfection activity.
Bile-acid-derived surfactants (43c) and (44) showed little transfection activity but
were good antibacterial agents. For example, the activity of (43c) (which contained one
serine and three lysine residues) was at least an order of magnitude higher than for
dioctadecyldimethylammonium bromide. Better antibacterial activity was observed for
Special gemini surfactants 271
(43c), with eight ammonium groups, than for (44), with two ammonium groups. The
authors suggest that the good activity is due to electrostatic interactions between the
cationic lysine residues and anionic phosphodiester groups in the bacterial membranes,
and between the hydrophobic cholyl group and the hydrophobic portion of the membrane
[96].
The aggregate state of bacteriophage T4 DNA upon addition of lysine-based surfactant
(45) was investigated using fluorescence microscopy. As surfactant is added to the DNA,
a random coil to globule transition occurs, via an intermediate coexistence region. The
concentration of surfactant at the boundaries of the coexistence region, given as a ratio of
surfactant to DNA charge, was independent of spacer length for surfactant (45) (with
spacer carbon number s=4, 6, 8). This was in contrast to 12-s-12 surfactants, for which
the concentration of the beginning of the coexistence region depended on the spacer
length. This dependence was explained in enthalpy and entropy terms, whereas the lack
of variation for surfactant (45) was attributed to the small variation in spacer lengths used
[100].
Cationic surfactants derived from the amino acid glycine with or without [(46a) and
(46b), respectively] a disulfide spacer were prepared. The authors suggest that these types
of surfactants are preferable to the m-s-m-type surfactants due to their potential to
biodegrade more easily. In addition, the principal breakdown product from these
surfactants is the well-known “soft” amphoteric surfactant N-dodecyl-N,N-
dimethylglycine. The cmc’s were 4.1× 10−5 M (46a) and 3.9×10−5 M (46b), although the
plots of surface tension versus concentration showed a slight dip around the cmc,
indicative of surface-active impurities. Nevertheless, these cmc values are extremely low
Gemini surfactants 272
Scheme 45–48
and may, in fact, be due to premicellization as found for arginine-based surfactants
studied by the same authors (see Chapter 5). The antimicrobial activities of (46a-46b)
against 19 micro-organisms were evaluated and found to be superior to
cetyltrimethylammonium bromide [101,102].
Surfactant (47) represents the only example of an anionic amino-acid-based gemini
surfactant. The backbone of (47) is composed of glycine and cysteine residues. The cmc
was determined to be approximately 0.1 mM. Diffusion coefficient measurements
Special gemini surfactants 273
revealed micelles of 3.5 nm in diameter for a 0.1 wt% solution. TEM also revealed
micellar aggregates, along with “globular strings” of approximately 40 nm in length.
Fibrillar aggregates with a width of 6–40 nm and length of over 10 µm formed upon
aging, which also exhibited a periodic “twist-like” feature [99].
Surfactants of type (48) and others were prepared by chemoenzymatic means. The
enzymatic step involved using a lipase enzyme to couple an α,ω-diol to protected amino
acids [45,103].
VII.
CONCLUSIONS
The majority of work on gemini surfactants has so far focused on the cationic class and
has helped develop a general understanding of the behavior of gemini surfactants.
Nevertheless, a large body of knowledge has been amassed on geminis other than
cationics, and much of the work on these “specialist” gemini surfactants has been
reviewed in this chapter. The interesting properties shown by theses novel surfactants
demonstrate the applicability of the gemini class to many different disciplines. Although
it is difficult to make general statements on the large range of work reviewed, the
nonionic geminis are of particular interest and will most likely represent a large
proportion of commercially available gemini surfactants in the future. Environmental
aspects, including benign chemoenzymatic syntheses, biocompatibility, and
biodegradability, will also be important.
REFERENCES
12. Tracy D.J.; Li R.; Dahanayake M.S.; Yang J. (assigned to Rhodia Inc., USA), US patent
5,945,393, August 31, 1999, filed November 7, 1997.
13. Tracy D.J.; Li R.; Dahanayake M.S.; Yang J. (assigned to Rhodia Inc., USA), US patent
6,204,297, March 20, 2001, filed November 26, 1996.
14. Tsubone K.; Nishio H.; Kusumam M. (assigned to Kanebo Ltd, Japan), Japanese patent
8,291,040, November 5, 1996, filed April 18, 1995.
15. Medina S.W. (assigned to Air Products, Chemicals, Inc.), US patent 5,650,543, July 22, 1997,
filed August 31, 1995.
16. Lassila K.R.; Uhrin P.A. (assigned to Air Products and Chemicals, Inc.), US patent 6,313,182,
November 6, 2001 filed May 4, 1999.
17. Castro, M.J.L.; Kovensky, J.; Cirelli, A.F. Tetrahedron, Lett. 1997, 38, 3995–3998.
18. Castro, M.J.L.; Kovensky, J.; Cirelli, A.F. Tetrahedron 1999, 55, 12,711–12,722.
19. Castro, M.J.L.; Kovensky, J.; Cirelli, A.F. Molecules 2000, 5, 608–609.
20. Castro, M.J.L.; Kovensky, J.; Cirelli, A.F. J. Carbohydr. Chem. 2000, 19, 1175–1184.
21. Castro, M.J.L.; Kovensky, J.; Cirelli, A.F. Langmuir 2002, 18, 2477–2482.
22. Castro, M.J.L.; Kovensky, J.; Cirelli, A.F. Tenside Surfact. Detergents 2002, 39, 28–30.
23. Eastoe, J.; Rogueda, P.; Harrison, B.J.; Howe, A.M.; Pitt, A.R. Langmuir 1994, 10, 4429–4433.
24. Briggs, C.B.A.; Newington, I.M.; Pitt, A.R. J. Chem. Soc. Chem. Commun., 1995, 379–380.
25. Eastoe, J.; Rogueda, P.; Howe, A.M.; Pitt, A.R.; Heenan, R.K. Langmuir 1996, 12, 2701–2705.
26. Briggs C.B.A.; Pitt A.R. (assigned to Eastman Kodak Co., USA; Kodak Ltd.), US patent
4,892,806, January 9, 1990, filed September 26, 1988.
27. Chierici, S.; Boullanger, P.; Marron-Brignone, L.; Morelis, R.M.; Coulet, P.R. Chem. Phys.
Lipids 1997, 87, 91–101.
28. Pestman, J.M.; Terpstra, R.; Stuart, M.C.A.; van Doren, H.A.; Brisson, A.; Kellogg, R.M.;
Engberts, J.B.F.N. Langmuir 1997, 13, 6857–6860.
29. van Doren, H.A.; Smits, E.; Pestman, J.M.; Engberts, J.B.F.N.; Kellogg, R.M. Chem. Soc. Rev.
2000, 29, 183–199.
30. Bergsma, M.; Fielden, M.L.; Engberts, J.B.F.N. J. Colloid Interface Sci. 2001, 243, 491–495.
31. Fielden, M.L.; Perrin, C; Kremer, A.; Bergsma, M.; Stuart, M.C.; Camilleri, P.; Engberts,
J.B.F.N. Eur. J. Biochem 2001, 268, 1269–1279.
32. van Eijk, M.C.P.; Bergsma, M.; Marrink, S.J. Eur. Phys. J. E 2002, 7, 317–324.
33. Camilleri P.; Engberts J.B.F.N.; Fielden M.L.; KremerA. (assigned to Smithkline Beecham
P.L.C., UK; University of Groningen), WO 0076954, December 21, 2000, filed June 16, 1999.
34. Fukunaga, K.; Mine, Y.; Samejima, K.; Yoshimoto, M.; Nakao, K.; Sugimura, Y. Biotechnol.
Lett. 2002, 24, 1157–1160.
35. Wathier, M.; Polidori, A.; Ruiz, K.; Fabiano, A.S.; Pucci, B. New J. Chem 2001, 25, 1588–
1599.
36. Zana, R. Langmuir 1996, 12, 1208–1211.
37. Fisicaro, E.; Pelizzetti, E.; Viscardi, G.; Quagliotto, P.L.; Trossarelli, L. Colloids Surfaces A
1994, 84, 59–70.
38. Wilk, K.A.; Syper, L.; Domagalska, B.W.; Komorek, U.; Maliszewska, I.; Gancarz, R. J.
Surfact. Detergents 2002, 5, 235–244.
39. Rajewska, A.; Komorek, U.; Wilk, K.A.; Islamov, A.H.; Kuklin, A.I. Small-angle neutron
scattering studies of micelles in aqueous solutions of nonionic Gemini surfactants. Liquid
Matter Conference, 2002, p. 183.
40. Menger, F.M.; Mbadugha, B.N.A. J. Am. Chem. Soc. 2001, 123, 875–885.
41. van Buuren, A.R.; Berendsen, H.J.C. Langmuir 1994, 10, 1703–1713.
42. Rose, J.D.; Maddry, J.A.; Comber, R.N.; Suling, W.J.; Wilson, L.N. Reynolds, R.C.
Carbohydr. Res. 2002, 337, 105–120.
43. Millqvist-Fureby, A.; Gao, C.; Vulfson, E.N. Biotechnol Bioeng 1998, 59, 747–753.
44. Gao, C.; Millqvist-Fureby, A.; Whitcombe, M.J.; Vulfson, E.N. J. Surfact. Detergents 1999, 2,
293–302.
Special gemini surfactants 275
I.
INTRODUCTION
Surfactants have been widely used in various fields from household and personal care
products in the consumer product industry to industrial applications in oil field chemistry,
agrochemistry, paint, and coating. Surfactant chemistry is also well established. A
mixture of surfactants is always used in applications so as to have an optimized
performance from the synergism of the mixture. Despite versatile applications in many
industries, surfactants seem old and have not received much attention for many years
from industries and scientific organizations. However, since around 1990, when the
gemini surfactant became a new category of the surfactant family, considerable interest
has been generated in this new class of surfactants from both industrial and academic
research organizations, which can be seen from some recent reviews [1–6]. In fact, Dow
Chemical, Sasol, and Air Product already have their gemini surfactants or gemini-
surfactant-based formulations in the market for special applications (Dowfax, Surfynol,
and Ceralution, respectively).
Gemini surfactants have unique structure and properties as pointed out in Chapter 1 in
this volume. Figure 1 provides schematic representations of the structures of a regular
(symmetric) and of a dissymmetric (heterogemini) gemini surfactants.
The research groups of Rosen, Zana, Menger, and Okahara have extensively studied
gemini surfactants [1–31] and their work and efforts have attracted a tremendous interest
about gemini surfactants, as can be seen with the numerous references on gemini
surfactants cited in this volume.
Because a gemini surfactant has two hydrophilic groups, it is more water soluble and
hard-water tolerant. Also, because a gemini surfactant has two hydrophobic groups, it
will more readily self-associate into micelles, as
Gemini surfactants 278
II.
EFFECT OF THE HYDROPHILIC GROUP OF GEMINI
SURFACTANTS ON PERFORMANCES
The hydrophilic group has been found to have an effect on water solubility, hard-water
tolerance, micelle forming ability, and other properties of gemini surfactants.
Structure-performance relationships in gemini surfactants 279
A.
Water Solubility, Krafft Point, and Cloud Point
Usually, the Krafft point, the temperature at which the solubility of an ionic surfactant
becomes equal to its critical micelle concentration (cmc), is used as an index of
measuring the water solubility of the surfactant. Below the Krafft point, the water
solubility of a surfactant is too small to enable the formation of micelles. Above the
Krafft point, the surfactant solubility generally increases dramatically with temperature.
As a rule, the lower the Krafft point, the higher the water solubility of the surfactant. In
contrast, the cloud point is the temperature at which an aqueous nonionic surfactant
solution gets hazy, followed by a phase separation. Thus, the higher the cloud point, the
higher the temperature range in which the nonionic surfactant remains solubilized. Both
the Krafft point and cloud point are of great practical significance.
Table 1 lists the Krafft points (strictly speaking, they are Krafft temperatures, which
are slightly higher than the real Krafft points) of gemini surfactants (1a-1n)
[ROCH2CH(OZ)CH2]2Y along with those of conventional surfactants. By comparison, it
is very clear that the gemini surfactants (1a-1n), regardless of the nature of ionic groups,
have lower Krafft points than the comparable conventional surfactants, implying that the
gemini surfactants (1a-1n) have high water solubility.
Table 2 summarizes the Krafft points of gemini surfactants sodium α-sulfonated fatty
acid polyethylene glycol diesters (2a-2h) [CmH2m+1 CH(SO3−
TABLE 1 Krafft Points of Gemini Surfactants (1a-
1n) and of Conventional Surfactants
Surfactant R Y Z Krafft point (°C) Ref.
− +
1a C10H21 OC2H4O SO3 Na <0 8
− +
1b C10H21 OC4H8O SO3 Na <0 8
− +
1c C10H21 p-OC6H4O SO3 Na <0 8
− +
1d C10H21 OC2H4O P(O)(O Na )2 <0 9
− +
1e C10H21 OC2H4O P(O)(O Na )(OH) <0 9
− +
1f C10H21 O (CH2)3SO3 Na <0 10
− +
1g C12H25 O (CH2)3SO3 Na <0 10
− +
1h C14H29 O (CH2)3SO3 Na <0 10
− +
1i C10H21 OC2H4O (CH2)3SO3 Na <0 10
− +
1j C10H21 OC2H4SC2H4O (CH2)3SO3 Na <0 15
− +
1k C10H21 OC2H4S(O)C2H4O (CH2)3SO3 Na <0 15
− +
1l C10H21 OC2H4S(O)2C2H4O (CH2)3SO3 Na <0 15
− +
1m C10H21 O CH2COO Na <0 14
Gemini surfactants 280
Na +)C(O)]2Y [34]. The gemini surfactants (2a-2h) showed high water solubility as
compared with the corresponding single-chain α-sulfonated fatty acid methyl ester.
Compound (3a-3d) [CmH2m+1]2C[CH2NHC(O)(CHOH)4CH2OH]2 [35] are nonionic
gemini surfactants derived from glucose, having two alkyl chains and two glucamide
groups. Compounds (3a-3d) are here referred to as Di(Cm-Glu), m=5,6, 7, and 8,
respectively. Di(C7-Glu) and Di(C8-Glu) were found to have a Krafft point around 34°C.
For all of the nonionic Di(Cm-Glu), there were no cloud points observed at least up to
100°C Singlechain N-methyl alkanoyl glucamides have been reported to have similar
phenomena with temperature, and the phenomena are explained in terms of hydrogen
bond and molecular crystallinity [36]. Di(C5-Glu) and Di(C6-Glu) were estimated to have
a Krafft point below 5 °C, because for both of them, no Krafft point was detected, at least
down to 5°C [35].
Nonionic gemini surfactants with a polyether hydrophilic head group were reported to
be extremely effective emulsifiers for an oil-in water emulsion at even low
concentrations, and their water solubility or cloud point can be manipulated by change
the number of oxyethylene/oxypropylene units [37].
Table 3 lists the solubility of cationic geminis [C12H25N+(CH3)2CH2C(O) NH]2Y, 2Cl−
with Y=(CH2)2SS(CH2)2 (4a) and (CH2)4 (4b) along with the data for
cetyltrimethylammonium bromide (CTAB) at 20 °C in the presence and absence of 10%
Structure-performance relationships in gemini surfactants 281
NaCl [38]. Clearly, surfactants (4a) and (4b) have a higher water solubility than CTAB in
the presence and absence of salt.
TABLE 3 Water Solubility of Compounds (4a–4b)
at 20°C
Surfactant In water In water+10% NaCl
4a ≤25% ≤20%
4b ≤20% ≤20%
CTAB ≤0.3% ≤0.15%
B.
Ability to Lower Surface Tension (γcmc)
Equation (1) describes the ability of a surfactant to lower the surface tension of a solvent:
(1)
In Eq. (1), γ0 is the surface tension of the solvent, Γmax is the surface excess
concentration, an index of the effectiveness of adsorption of the surfactant at the
interface, n is a constant that depends on the surfactant, and C20 is the surfactant
concentration in the bulk phase required to reduce the surface tension of the solvent by 20
mN/m. It is obvious that the γcmc value depends on the effectiveness of adsorption of the
surfactant Γmax and the cmc/C20 ratio. Both Γmax and the cmc/C20 ratio strongly depend on
Gemini surfactants 282
the surfactant structure. The γcmc values for gemini surfactants with the same hydrophobic
group and spacer are summarized in Table 4. It is evident that the ability in lowering the
surface tension (γcmc) of ionic gemini surfactants, except the taurine-based compound, is
higher than that of the corresponding conventional ionic surfactants. Regarding the effect
of the hydrophilic group, it can be concluded by comparison of the data that a too bulky
hydrophilic group, which can hinder the surfactant adsorption and effective packing at
the air-solution interface, can have a negative effect on lowering the surface tension. A
similar effect of an increase in the size of the hydrophilic group, without a significant
change in its nature, on the γcmc values has been reported [41]. This effect perhaps results
from a smaller Γmax and less or little change in the cmc/ C20 value when the size of the
hydrophilic group increases.
C.
Foaming Properties, Lime-Soap Dispersing Ability, and Calcium
Stability
Foaming properties, lime-soap dispersing ability (LSDR), and calcium stability of
surfactants are very important in practical applications. Foam control,
TABLE 4 Values of γcmc of Gemini Surfactants
with Different Hydrophilic Groups at 20 °C
Surfactant Y z γcmc (mN/m) Ref.
− +
OC2H4O SO3 Na 27.0 8
− +
OC2H4O P(O)(O Na )2(OH)2 32.5 9
− +
OC2H4O (CH2)3SO3 Na 30.0 10
− +
OC2H4O CH2COO Na 33.0 14
− +
OC2H4O C2H4SO3 Na 37.0 12
C12H25SO3−Na+ 39.5 10
− +
C12H25OSO3 Na 39.0 8
− +
C11H23COO Na 39.5 14
− +
C12H25OP(O)(O Na )2 37.5 33
high lime-soap dispersing ability, and high calcium stability are always desirable in
applications.
Table 5 summarizes the foaming properties, lime-soap dispersing ability, and calcium
stability of homologous gemini surfactants with oxyethylene as a spacer and C10H21 and
C12H21 alkyl group for surfactants (1) and surfactants (2). Data for conventional
surfactants are also given for the purpose of comparison. The LSDR value is the amount
of a surfactant required to disperse a certain amount of lime-soap, namely the lower the
LSDR value, the higher the lime-soap dispersing ability.
For the gemini surfactants listed above (la, 1e, 1i; and In), except (1d), (2a) and (2d),
the initial foam heights are comparable to that of C12H25OSO3− Na+, which is a well-
Structure-performance relationships in gemini surfactants 283
known good foaming agent. Compounds (1d) and (1e) are both phosphate salts. The
tetrasodium salt (1d) showed low foaming properties. In contrast to this, the disodium
salt (1e) showed high foaming and foam stability. The tetrasodium salt has more
electrostatic repulsion between the surfactant molecules adsorbed onto the foam surface,
resulting in less good foaming properties.
The sodium α-sulfonated fatty acid polyethylene glycol diesters (2a) and (2d) showed
low foam properties.
All of the gemini surfactants in Table 5 have much lower LSDR values than
conventional surfactants, indicating that they are very efficient lime-soap dispersing
agents. Also gemini surfactants, even dicarboxylate geminis, have high calcium stability.
Thereby, the gemini surfactants are applicable to many uses in hard water without
needing additives.
TABLE 5 Foaming Properties, Lime-Soap
Dispersing Ability, and Calcium Stability
Foam Volume (mL)a
Surfactant Initial 5 min LSDR Calcium stability (ppm) Ref.
1a 250 0 5.8 — 8
1d <10 0 — — 9
1e 240 230 — — 9
1i 255 225 6.3 >5500 10
1n 250 250 — 650 14
2a 80 — — >1800 34
2d 68 — — >1800 34
− +
C12H25OSO3 Na 240 240 30 1080 8, 14
− +
C12H25SO3 Na 215 130 94 212 10
− +
C11H23COO Na 200 200 — 250 14
a
Measured with a 0.1 wt% aqueous surfactant solution.
van Zon et al. also reported that the gemini surfactants derived from 2-hydroxy-
alkylsulfonate had much higher calcium stability than the corre sponding α-olefin
sulfonate [42].
D.
Miscellaneous
Gemini surfactants with the structure [R-C6H4-SO3−M+]2Y, Y=N, O, and S, are claimed
to demonstrate improved stain removal from skin and clothing as compared with sodium
dioctylsulfosuccinate and sodium sulfate [43]. The stain-removal efficacy is affected by
Gemini surfactants 284
the counterions, and the geminis with N-aminoethylpiperazine as a counterion show high
stain-removal performance for inks and coloring materials.
Anionic geminis with the structure [C8H17C6H4O(C2H4O)xSO3−M+]2CH2 were tested
by the Eytex protocol and claimed as mild surfactants, even milder than sodium
laurylethersulfate [44]. It is noteworthy that unlike conventional surfactants, the number
of C2H4O units between the phenyl group and the charged head group has relatively little
effect on skin irritation [44]. However, wetting properties for a cotton skein depend on
the number of oxyethylene units x. An increase of x results in an increase in Draves
skein-wetting times.
III.
EFFECT OF THE HYDROPHOBIC GROUP OF GEMINI
SURFACTANTS ON PERFORMANCES
A.
Water Solubility
In general, the water solubility of a gemini surfactant decreases as the hydrophobic
character of the molecule increases; namely the Krafft point goes up for ionic geminis,
and the cloud point goes down for nonionic geminis upon increasing alkyl chain length.
This rule is also true for conventional surfactants. The data listed in Table 2 clearly show
that the Krafft point of the sodium α-sulfonated fatty acid ethylene glycol diesters
increases from below 0°C to 23°C when the fatty acid alkyl chain changes from C14 to
C16 and C18 [34].
Also, for the cationic gemini surfactants [Cl−, R1R2N+CH3]2CH2CH-(OH)CH2 (5) and
[Cl−, R1N+(CH3)2]2CH2CH2CH2 (6) (see structures above), with R=C18H37, the Krafft
points are close to 40°C, whereas when the alkyl chain is shorter than C18, their Krafft
points are below 0°C [39].
A study of cationic gemini surfactants with an acetylenic spacer reported that the
water solubility decreases as the alkyl chain length increases [20]. Ikeda et al. [45] have
reported that gemini bis(ester ammonium) dichlorides with butenylene spacer and alkyl
chains up to C14H29 have Krafft points below 0°C. The same geminis with C16 and C18
alkyl chains have Krafft points of 7°C and 46°C, respectively.
B.
The C20 Value
The C20 values of gemini surfactants are two to three orders of magnitude smaller than
those of the corresponding conventional surfactants [5]. Similar to the effect of the alkyl
chain length on the cmc value, the C20 value also decreases with an increase in the alkyl
chain length. The introduction of a double bond into the alkyl chain has been reported to
make a negative effect on the efficiency of adsorption on the surface [46], resulting in an
increase in the C20 values.
The value of pC20=−log C20 measures the efficiency of adsorption of the surfactant at
the air-water interface. For almost all of the gemini surfactants with an alkyl chain shorter
Structure-performance relationships in gemini surfactants 285
than C16H33 [strictly speaking, this depends on the molecular environment (i.e., the
chemical structure of the gemini surfactant, temperature, absence, and presence of
electrolytes)] regardless of the charge of the hydrophilic group, the pC20 value increases
linearly with the alkyl chain carbon number m. On the other hand, it has been reported
that some gemini surfactants with an alkyl chain longer than C16H33 show an abnormal
behavior [40,47–49], resulting in departures from the linear relationship in the plots of
pC20 versus m. This phenomenon has been attributed to premicellar aggregation when the
alkyl chain is C16H33 and longer [47,48] (see Chapter 5, Section III).
Table 6 lists the values of pC20 and of cmc/C20 for the cationic geminis
[CmH2m+1(CH3)2N+, Br−]2CH2CH2OCH2CH2 (7a–7d) in 0.1 M NaCl at 25°C [47]. It is
apparent that the pC20 values increase with the alkyl chain up to C14, and the high
homologs with C16 and C18 alkyl groups deviate from this trend and result in a decrease in
the pC20 values.
The cmc/C20 ratio is a measure of the surfactant preference for adsorption relative to
micellization. The large decrease in the cmc/C20 ratios when m reaches 16 may reflect
multilayer formation, as suggested by the large slope in the plots of surface tension (γ)
versus log C [47].
TABLE 6 Values of pC20 and of the cmc/C20 ratio
for the Geminis [CmH2m+1(CH3)2N+, Br−]2
CH2CH2OCH2CH2 in 0.1 M NaCl at 25°C
Surfactant pC20 cmc/C20
7a (m=12) 6.0 27
7b (m=14) 7.3 24
7c (m=16) 6.5 3.0
7d (m=18) 6.4 1.8
Source: Ref. 47.
C.
Ability to Lower Surface Tension (γcmc)
According to Eq. (1), the alkyl chain length of the hydrophobic group is expected to have
a small effect on the γcmc value for ionic gemini surfactants.
For the two series of bis(sulfonate) geminis [ROCH2CH(OC3H6SO3− Na+)CH2]2Y with
Y=O or OCH2CH2O, the γcmc values depend little on the alkyl chain length and increases
slightly with the alkyl chain length when it exceeds C12H25 [10,50]. For cationic gemini
surfactants with an acetylenic spacer [20], the alkyl chain length has only a small effect
Gemini surfactants 286
on the value of γcmc, which starts to increase for alkyl chains longer than C14 or C16. It is
also noteworthy that the cationic geminis with a heterocyclic group show γcmc values
lower than the corresponding geminis with an aliphatic alkyl chain. The γcmc values of
cationic gemini series (5) and (6) have been reported to increase slightly with alkyl chain
length [39]. A similar effect of the alkyl chain length on γcmc values was reported for
other geminis [18,19,40,45,49]. The observed increase in γcmc value with increasing alkyl
chain length of ionic gemini surfactants is explained in terms of premicellar aggregation,
self-coiling of the long alkyl chain. Unsaturation in the alkyl chain caused an increase in
the γcmc values for bis(sulfonate) geminis [46].
For nonionic gemini surfactants, according to Eq. (1), an increase in the alkyl chain
length is expected to result in a reduction in the γcmc value because of an increase in Γmax
value with increasing alkyl chain length. Table 8 lists the γcmc values and area per gemini
surfactant molecule at the surface (amin) of nonionic gemini (3a–3d) [35].
TABLE 7 Interfacial Properties of
[CmH2m+1(CH3)2N+ ,Br−]2CH2CH(OH)CH2 in 0.1 M
NaCl at 25°C
Surfactant System pC30 cmc/C20 γcmc (mN/m)
8a (m=10) Hexadecane/0.1 M NaCl 4.60 20 6.4
8b (m=12) Hexadecane/0.1 M NaCl 6.64 61 3.4
8c (m=14) Hexadecane/0.1 M NaCl 6.84 19 2.7
8d (m=16) Hexadecane/0.1 M NaCl 5.28 6.1 4.7
Source: Ref. 47.
For bis(taurine) gemini surfactants, which have a very large and bulky hydrophilic group,
the γcmc values were found to decrease with an increase in the alkyl chain length from
octyl to decyl and to dodecyl because of the increased effectiveness of adsorption with
alkyl chain length [12].
Structure-performance relationships in gemini surfactants 287
D.
Foaming Properties and Lime-Soap Dispersing Ability
For disulfonate geminis of structure [ROCH2CH(OC3H6SO3−Na+)CH2]2O, with
R=C10H21, C12H25 and C14H29, the initial foam height is almost independent of the alkyl
chain length, but the foam stability drops as the chain length is increased [10]. On the
other hand, bis(taurine) geminis of structure [RN(C2H4SO3−Na+)CH2CH(OH)CH2OCH2]2
show improved foaming ability and foam stability as the chain length is increased [12].
The same behavior has been noted for geminis with three alkyl chains [11,13].
Cationic geminis with the structure [RN+(CH3)2,Cl−]Y where Y=CH2 CH(OH)CH2 or
(CH2)3, were observed to have a maximum foaming ability and foam stability with
R=C12H25 and C14H29 [39]. When R is shorter than C12H25 or longer than C14H29, the
geminis showed poor foaming properties. A similar effect of alkyl chain length on
foaming properties was observed for cationic geminis [RC(O)O(CH2)nN+(CH3)2,Cl−]2Y,
with Y=CH2CH(OH) CH2 or trans-CH2CH=CHCH2 [45,51]. It is noteworthy that these
cationic geminis show higher foaming properties than conventional cationic surfactants
and are even comparable to that of conventional anionic surfactants.
For disulfonate geminis of structure [ROCH2CH(OC3H6SO3−Na+) CH2]2O, the lime-
soap dispersing ability is much higher than that for C12 H25SO4−Na+ and C12H25SO3−Na+,
and it decreases a little as the alkyl chain length increases [10]. For the triple-chain
geminis [11,13], the lime-soap dispersing ability depends a little on the alkyl chain
length. In addition, the presence of an acyl group enhances the lime-soap dispersing
ability.
The wetting ability decreases when the alkyl chain is longer than C10H21 [11,52].
Geminis with short and branched hydrophobic groups and with a short spacer show high
wetting ability [11,13,14].
E.
Removal of Pollutants from Aqueous Media
Adsorbed layers of surfactants on solid particles constitute hydrophobic domains which
are able to solubilize/adsolubilize water-insoluble materials.
The adsorption studies of 2-naphthol and 4-chlorophenol as model pollutants onto
adsorbents treated with conventional and cationic gemini surfactants showed that the
geminis are more efficient and more effective at removing those pollutants from aqueous
media [53–55]. The ratio of maximum amount of adsorbed pollutant to the amount of
adsorbed surfactant on the adsorbent increases with the alkyl chain length [53,54] as well
as with the number of alkyl chains of the surfactant [55], showing that gemini surfactants
are more effective than conventional surfactants. The maximum adsorption of gemini
surfactants onto the adsorbent takes place at a lower concentration than for conventional
surfactant in the solution phase, implying that the gemini surfactants are more efficient at
removing the organic pollutants from aqueous media [53,54].
In the study of 2-naphthol adsolubilization on surfactant-coated silica surface, it was
concluded that gemini surfactants adsorb strongly onto silica through electrostatic
Gemini surfactants 288
interaction and hydrophobic interaction between alkyl chains of the gemini surfactants
[55].
The anionic gemini surfactant disodium didecyldiphenylether disulfonate (DADS) is
reported to be more efficient and more effective than the conventional anionics in the
removal of 2-naphthol from the aqueous phase when using limestone as an adsorbent
[54]. This anionic gemini is reported to adsorb as a single layer, with its second
hydrophilic group oriented toward the aqueous phase. In contrast, the conventional
anionics adsorb onto limestone to form a bilayer with the hydrophilic groups of the
second adsorbed layer oriented toward the aqueous phase.
F.
Miscellaneous
Use of the sugar-derived nonionic geminis [R′NHCH2CH2N(R)CH2]2, with R=acyl group
and R′=lactobionyl, is claimed to result in unexpectedly reduced mean particle size and to
control the particle size growth during the terminal sterilization of the nanocrystal
formulation when making drug dispersions by wet-milling techniques, giving an average
particle size of less than 250 nm [56]. Use of the gemini with R=Ph(CH2)10C(O) yielded
particles of average size of less than 150 nm, whereas use of the gemini with R=
C12H25HNC(O) yielded a dispersion with particle size larger than 150 nm.
The nonionic geminis (CmH2m+1)2C[CH2NHC(O)(CHOH)xCH2OH]2 were tested as a
coating aid for making photographic materials [57]. These geminis are claimed to give a
film almost free of repellencies and craters, whereas the corresponding single-chain 2-
ethylhexyl gluconamide gives a film covered with a larger number of repellencies [57].
Also, an increase of m from 6 to 7 enhanced the efficiency as a coating aid, resulting in
lowering the concentration required to give a film free of repellencies and craters.
The geminis [RCH(SO3−Na+)C(O)]2Y are claimed to induce much less skin irritation
than conventional anionic surfactants such as sodium dodecyl sulfate and potassium
myristate [52]. Also different from that for conventional anionic surfactants, an increase
of alkyl chain length from C12 to C16 brought no significant effect on skin irritation [52].
The alkyl chain length affects the wetting which is a maximum for surfactants with a
C10—C12 alkyl chain [52].
The cationic geminis [RN+(CH3)2,Cl−]2CH2CH(OH)CH2, tested for substantivity to
hair by use of the Rubine dye test, are claimed to be very substantive to hair and thus
would be a very effective hair conditioner when R is stearyl (C18) or docosanyl (C22) [58].
The substantivity to hair depends on the alkyl chain length, and an increase in the alkyl
chain length enhances the substantivity.
Structure-performance relationships in gemini surfactants 289
IV.
EFFECT OF THE SPACER GROUP OF GEMINI SURFACTANTS
ON PERFORMANCES
A.
Ability to Lower Surface Tension (γcmc), Area per Molecule, pC20
Value, and cmc/C20 Ratio
The structure of the spacer affects the packing of gemini surfactants at the air-water
interface. The gemini [ROCH2CH(OCH2COO−Na+)CH2]2Y, with Y=O(C2H4O)x, x=0–3,
and (CH2)4 [14], show γcmc values that increase with x. The gemini with an oxygen atom
as a spacer (x=0) shows the lowest γcmc value. When the spacer between the two head
groups is small or hydrophilic, the area per hydrophobic chain at the aqueous solution-air
interface has been found to be smaller with geminis than with comparable conventional
surfactants. In the above series of geminis, the area per molecule at the air–0.1 M NaCl
solution interface at 20°C (pH 11) increases in the order O<
OC2H4O<O(C2H4O)2<O(C2H4O)3. However, even for the spacer O(C2 H4O)3, the area per
hydrophobic chain is smaller than for C11H23COO−Na+ under the same conditions. The
C20 values of the geminis are lower than those for the corresponding conventional
surfactants by about two orders of magnitude. This indicates that the efficiency of
adsorption of surfactants is increased greatly by the introduction of the second
hydrophobic group into the molecule. The pC20 values decrease with increasing value of
x. This may be due to the increase in the hydrophilic character of the geminis with x. For
these geminis, an increase in x results in a monotonic decrease in the cmc/C20 ratio. It is
noteworthy that the gemini with a (CH2)4 spacer showed a large cmc/C20 ratio among the
series of the above gemini, indicating the steric inhibition of micellization by this
hydrophobic (CH2)4 spacer.
Concerning the effect of the hydrophobic spacer on γcmc values, it has been reported
for the geminis [CmH2m+1(CH3)2N+,Br−]2(CH2)s that the γcmc values slightly increase with
the spacer carbon number s [59].
For the gemini surfactants [C10H21OCH2CH(OZ)CH2]2Y, where Y is (Z)-
OCH2CH=CHCH2O, (E)-OCH2CH=CHCH2O, OC2H4O, and OC4H8O, and
Z=(CH2)3SO3−Na+ or CH2COO−Na+, the geminis with a double bond in the spacer have
smaller γcmc values than the geminis with a comparable but saturated spacer [46]. It is
noteworthy that the replacement of a C-C bond by a C=C bond in the spacer results in an
increase in values of pC20 and cmc/C20, facilitating the gemini adsorption at the aqueous
solution-air interface.
For the geminis [C12H25(CH3)2N+,Cl−]2, YwithY=CH2CH(OH)CH2, (CH2)3,
CH2CH(OH)CH2CH2, and CH2CH2OCH2CH2, the spacer has a relatively small effect on
the γcmc values and pC20 values [39].
Geminis with the structure [C10H21OCH2CH(OC3H6SO3−Na+)CH2]2Y, where
Y=OCH2CH2SCH2CH2O, OCH2CH2S(O)CH2CH2O, and OCH2 CH2S(O)2CH2CH2O,
show that the spacer affects the values of γcmc and pC20 because of the difference in
polarity and bulkiness of the connecting group [15].
Gemini surfactants 290
B.
Foaming Properties, Lime-Soap Dispersing Ability, and Wetting
Properties
For geminis with the structure [C10H21OCH2CH(OC3H6SO3−Na+)CH2]2Y, with a short
spacer (Y=O or OC2H4O), the initial foam volume and foam stability are significantly
higher than for conventional anionic surfactants. For Y=(OC2H4O)x, the initial foam
volume decreases slightly as x is increased from 1 to 3, but the foam stability decreases
substantially [10], reflecting the known decrease in foam stability as the area per
molecule at the air-solution interface increases. For the geminis with structure
[C10H21OCH2CH(OCH2-COO−Na+)CH2]2Y, a similar effect of the spacer Y on foam
properties has been reported [14].
For the geminis [C12H25(CH3)2N+,Cl−]2Y, with Y=CH2CH(OH)CH2, (CH2)3,
CH2CH(OH)CH2CH2, and CH2CH2OCH2CH2, the spacer Y has a marked effect on foam
stability [39]. This again indicates that the spacer has more effect on foam stability than
on foaming ability.
The geminis with structure [C10H21OCH2CH(OZ)CH2]2Y, where Z = SO3−Na+ or
C3H6SO3−Na+, have a higher lime-soap dispersing ability than the conventional
surfactants, regardless of the structure of the spacer Y. No relationship between lime-soap
dispersing ability and the structure of the spacer was clarified. However, these geminis
were found to be very stable in the presence of Ca2+ and Mg2+. The presence of
oxyethylene units in the spacer was found to enhance the calcium stability of these
geminis: The higher the number of oxyethylene units in the spacer, the better the stability
for Ca2+and Mg2+ [14].
Geminis in which the alkyl chains are short and branched and the spacer between the
two head groups is short show excellent wetting properties. The acetylenic glycols
(Surfynol) and their ethoxylates are used as low-foam wetting agents.
The nonionic geminis with the structure [RN[(C2H4O)xH]C(O)]2Y, where Y=C2H4 or
CH=CH and R=2-ethylhexyl, have been reported to be excellent hydrophobic soil-
wetting and soil-rewetting agents [60]. The value of x affects the wetting properties, and
the most effective wetting agent was found for x=4.
For geminis with the structure [C10H21OCH2CH(OZ)CH2]2Y, with Z = SO3−Na+ and
CH2CO2−Na+, the Draves skein wetting times at 0.1 wt% concentration depends on the
structure of the spacer. The shorter spacer O gave a short wetting time when Z was
CH2COO−Na+ [14], whereas the (CH2)4 spacer gave the shortest wetting time when Z was
SO3−Na+ [8].
For the geminis [RCH(SO3−Na+)C(O)]Y, where Y=O(C2H4O)x, the value of x affects
the wetting properties, and it is claimed that a low value of x is desirable for better
wetting properties. A maximum in wetting properties occurs at around x=4 [52].
Structure-performance relationships in gemini surfactants 291
C.
Miscellaneous Studies
1.
Adsorption
The effect of the spacer on the dynamics of adsorption of cationic gemini surfactants has
been studied [61]. It has been reported for all of the geminis studied that about 54–69%
of the interface is covered at the end of the induction period. The induction time of the
geminis with flexible hydrophilic spacers is larger than that of comparable geminis with a
rigid hydrophobic spacer under similar conditions. This is probably because the former
can form larger linear aggregates through chain-chain interaction. This was confirmed by
viscosity measurements.
A study on the dynamics of adsorption of anionic geminis surfactants showed that
adsorption at the air-water interface is diffusion controlled and that the value of the
dynamic adsorption parameter increases with an increase in the hydrophobic character of
the surfactant and decreases with an increase in the length of the polyoxyethylene chains
in the spacer [62]. It is apparent that the molecular size and shape of a surfactant play
major roles in the mechanism of adsorption.
Studies on the adsorption behavior of gemini diols [63] and triols [64] by measuring
pressure-area isotherms show the following relationship between the surfactant structure
and adsorption behavior at the air-water interface. A more tightly packed monolayer is
formed as the length of alkyl chain increases and as the length of the hydrophobic spacer
is shortened. A C=C bond in the alkyl chain results in loose packing of the monolayer.
No influence of chirality in the molecule on the adsorption behavior is observed. Packing
of alkyl chains of triple-chain diol was tighter than that of the corresponding double-
chain diol with the same alkyl chain length. The spacer greatly affects the pressure-area
isotherms.
2.
Gelling
Cationic geminis [R1N+H(R2)]2Y,Ar(COO−)2, where Y is an alkylene group (CH2)s, are
claimed to form a viscoelastic gel in aqueous solution [65]. The length of the spacer
significantly affects the gel formation. Preferably, the alkylene spacer should contain less
than six carbon atoms. A maximum gelling performance is found when Y=(CH2)3. The
carboxylic acid groups must be preferably attached to alternate carbons of the aromatic
ring Ar. The position of carboxylic acid groups on the ring affects the gelling
performance. The preferred aromatic dicarboxylic acid is isophthalic acid.
3.
Biodegradation
Cationic geminis derived from L-arginine have been reported to be readily biodegradable
and environmentally compatible [66]. The anionic gemini surfactants
Gemini surfactants 292
4.
Emulsions, Microemulsions
Gemini diols with the structure [C10H21OCH2CH(OH)CH2]2Y, where Y=
OCH2CH2OCH2CH2O, OCH2CH2SCH2CH2O, OCH2CH2S(O)CH2CH2O, and
OCH2CH2S(O)2CH2CH2O, were evaluated as emulsifiers for oil-in-water emulsions,
using liquid paraffin as an oil phase along with Span-40 as a control [15]. The stability of
the emulsion decreased in the sequence Y= OCH2CH2S(O)2CH2CH2O>Span-
40>>OCH2CH2S(O)CH2CH2O>OC H2CH2OCH2CH2O≥OCH2CH2SCH2CH2O.
From a phase-behavior study of water-surfactant-styrene ternary system, cationic
geminis [C12H25(CH3)2N+,Br−]2(CH2)s were found to be superior in solubilizing styrene in
water to conventional single-chain surfactants and in forming a single-phase oil-in-water
microemulsion in the temperature range from 25 °C to 60 °C [67]. The size of the
microemulsion region is strongly dependent on the spacer chain length. For s=2, only a
very small microemulsion region was formed, whereas for s=4, 6, 8, and 10, the extent of
the microemulsion region gradually increased. For s=12, the microemulsion region was
more temperature dependent. Studies on the phase behavior and microemulsion structure
of the cationic geminis with a hydrophilic oligo(oxyethylene) spacer
[C12H25(CH3)2N+,Br−]2(CH2)2(OCH2CH2)x showed that the geminis can form a stable
microemulsion phase with styrene at both 25 °C and 60 °C [68]. The size of the
microemulsion region greatly depends on the number x. The range of the microemulsion
phase is the largest when x=1 and decreases with increasing x.
The anionic gemini didodecyldiphenylether disulfonate (C12DADS) was reported to
form a microemulsion phase in the water-gemini-cosolvent-toluene system [69]. A phase-
behavior study of anionic gemini sodium 1,2-bis(N-dodecanoyl β-alanate)-N-ethane in a
water-surfactant-oil-hexanol system showed a very narrow microemulsion region and a
Structure-performance relationships in gemini surfactants 293
large domain for the lamellar phase. The replacement of hexanol by butanol resulted in a
broad microemulsion region [70].
V.
CONCLUSIONS
This chapter reviewed the relationships between performances and structure of gemini
surfactants. The gemini surfactants indeed have performances that conventional
surfactants or even their mixtures cannot achieve. Gemini surfactants have much higher
hard-water tolerance and lime-soap dispersing ability than conventional surfactants. The
wetting properties can be maximized by manipulating the structure of gemini surfactants.
Foam properties can be controlled with gemini surfactants by varying the hydrophilic
group, the hydrophobic group, and the spacer. The enhanced hydrophobic interaction,
while keeping sufficient water solubility, makes gemini surfactants more efficient and
effective in many applications. The tighter packing of the hydrophobic groups of gemini
surfactants results in a more cohesive, more stable film at interfaces, which can bring
greater emulsion stability and greater foam stability. The aggregate geometry at solid-
liquid interfaces can be controlled by changing the gemini surfactant structure. This
means that the use of gemini surfactants as a template can offer great opportunities for
material scientists to develop new materials with desirable surface charge density, surface
area, pore size, and other preferred properties. With gemini surfactants, hydrophobic
groups, hydrophilic groups, spacers, or even counterions can each be systematically
varied to study the structural effect on performances. Nevertheless, more studies are
needed to fully understand the structure-performance relationships in gemini surfactant
such that many possibilities of gemini surfactants can be explored and used.
REFERENCES
14. Zhu, Y.-P.; Masuyama, A.; Kobata, Y.; Nakatsuji, Y.; Okahara, M.; Rosen, M.J. J. Colloid
Interface Sci. 1993, 158, 40.
15. Zhu, Y.-P.; Masuyama, A.; Nakatsuji, Y.; Okahara, M. J. Jpn. Oil Chem. Soc. (Yukagaku)
1993, 42, 86.
16. Zhu, Y.-P.; Ishihara, K.; Masuyama, A.; Naktsui, Y.; Okahara, M. J. Jpn. Oil Chem. Soc.
(Yukagaku) 1993, 42, 161.
17. Rosen, M.J.; Gao, T.; Nakatsuji, Y.; Masuyama, A. Colloid Surfaces A 1994, 88, 1.
18. Menger, F.M.; Littau, C.A. J. Am. Chem. Soc. 1991, 113, 1451.
19. Menger, F.M.; Littau, C.A. J. Am. Chem. Soc 1993, 115, 10083.
20. Menger, F.M.; Keiper, J.S.; Azov, V. Langmuir 2000, 16, 2062.
21. Menger, F.M.; Migulin, V.A. J. Org. Chem. 1999, 64, 8196.
22. Rosen, M.J.; Zhu, Z.H.; Hua, X.Y J. Am. Oil Chem. Soc. 1992, 69, 30.
23. Zana, R.; Benrraou, M.; Rueff, R. Langmuir 1991, 7, 1072.
24. Zana, R.; Talmon, Y. Nature 1993, 362, 228.
25. Peresypkin, A.V.; Menger, F.M. Org. Lett 1999, 1, 1347.
26. Danino, D.; Talmon, Y.; Zana, R. J. Colloid Interface Sci 1997, 185, 84.
27. Danino, D.; Talmon, Y.; Zana, R. Langmuir 1995, 11, 1448.
28. Alami, E.; Levy, H.; Zana, R.; Skoulios, A. Langmuir 1993, 9, 940.
29. Alami, E.; Beinert, G.; Mari, P.; Zana, R. Langmuir 1993, 9, 1465.
30. Zana, R.; In, M.; Levy, H.; Duportail, G. Langmuir 1997, 13, 5552.
31. Rosen, M.J.; Zhu, Z.H.; Gao, T. J. Colloid Interface Sci 1993, 157, 254.
32. Rosen, M.J. Surfactants and Interfacial Phenomena, 2nd Ed.; Wiley: New York, 1989; 216 pp.
33. Imokawa, G.; Tsutusmi, H.; Kurosaki, T. J. Am. Oil Chem. Soc. 1978, 55, 839.
34. Okano, T.; Egawa, N.; Fujiwa, M.; Fukuda, M. J. Am. Oil Chem. Soc. 1996, 73, 31.
35. Eastoe, J.; Rogueda, P.; Howe, A.; Pitt, A.; Heenan, R. Langmuir 1996, 12,2701.
36. Zhu, Y.-P.; Rosen, M.J.; Vinson, P.K.; Morrall, S.W. J. Surfact. Detergents 1999, 2, 357.
37. Tracy, D.; Li, R.; Dahanayake, M.; Yang, J.; US patent 5,811,384, 1998.
38. Diz, M.; Manresa, A.; Pinazo, A.; Erra, P.; Infante, M.R. J. Chem. Soc. Perkin Trans. 2 1994,
1871.
39. Kim, T.; Kida, T.; Nakatsuji, Y.; Hirao, T.; Ikeda, I. J. Am. Oil Chem. Soc. 1996, 73, 907.
40. Rosen, M.J.; Liu, L. J. Am. Oil Chem. Soc. 1996, 73, 885.
41. Venable, R.L.; Nauman, R.V. J. Phys.Chem 1964, 68, 3498.
42. van Zon, A.; Bouman, J.T.; Deuling, H.H.; Karaborni, S.; Karthaeuser, J.; Mensen, H.; van Os,
N.M.; Raney, K.H. Tenside Surf. Det. 1999, 36, 84.
43. Kaiser, R.J.; US patent 5,487,778, 1996.
44. Tracy, D.; Li, R.; Ricca, J.M.; US patent 5, 710, 121, 1998.
45. Tatusmi, T.; Zhang, W.; Kida, T.; Nakatsuji, Y.; Ono, D.; Takeda, T.; Ikeda, I. J. Surfact.
Detergents 2001, 4, 279.
46. Masuyama, A.; Endo, C..; Takeda, S.; Nojima, M.; Ono, D.; Takeda, T. Langmuir 2000, 16,
368.
47. Rosen, M.J.; Mathias, J.H.; Davenport, L. Langmuir 1999, 15, 7340.
48. Mathias, J.H.; Rosen, M.J.; Davenport, L. Langmuir 2001, 17, 6148.
49. Song, L.D.; Rosen, M.J. Langmuir 1996, 12, 1149.
50. Kondo, Y.; Yoshino, N.; Isoda, T.; Abe, M.; Masuyama, A.; Nakatsuji, Y. J. Jpn. Oil Chem.
Soc.(Yukagaku) 1995, 44, 10.
51. Tatusmi, T.; Zhang, W.; Kida, T.; Nakatsuji, Y.; Ono, D.; Takeda, T.; Ikeda, I. J. Surfact.
Detergents 2000, 3, 167.
52. Okano, T.; Fukuda, M.; Tanabe, J.; Ono, M.; Akabane, Y.; Takahashi, H.; Egawa, N.; Sakatani,
T.; Kanao, H.; Yoneyama, Y.; US patent 5,681,803, 1997.
53. Li, F.; Rosen, M.J. J. Colloid Interface Sci. 2000, 224, 265.
54. Rosen, M.J.; Li, F. J. Colloid Interface Sci. 2001, 234, 418.
55. Esumi, K.; Goino, M.; Koide, Y. J. Colloid Interface Sci. 1996, 183, 539.
Structure-performance relationships in gemini surfactants 295
I.
INTRODUCTION
II.
CLEANING AGENTS AND DETERGENTS
monoalkyldiphehylether disulfonate;
DAMS=dialkyldiphenylether
monosulfonate,
DADS=dialkyldiphenylether
disulfonate. (Reproduced from Refs. 6
and 7 with permission.)
C10DADS-C12EO3 have superior wetting action and better solubilization [8]. The washing
and cleaning power of the surfactant is enhanced.
A patent [9] reports that detergent formulations comprising a conventional surfactant,
a gemini surfactant, and a polymeric nonionic soil release agent can be used as surfactant
additive packages, detergents, and fabric softeners. The blends not only promote the
deposition of soil release agents but also enhance soil removal, general detergency, and
soil antiredeposition. Here, the low cmc of the system due to the presence of gemini
surfactants causes better solubilization, increases detergency at lower surfactant levels,
and unexpectedly enhances the deposition of the soil-release polymer. Soil-removal
agents adhered to the laundered fabric much better than when mixed only with
conventional surfactants. Detergent formulations containing gemini surfactants were
tested for effectiveness by measuring fabric reflectance (R). A higher reflectance means a
cleaner fabric. The soil release performance was evaluated using
(1)
The results are listed in Table 1 [9]. Samples 2–4 show an increase in reflectance starting
from as little as 0.5% gemini surfactant MBOP on cotton/ polyester. The reflectance
dramatically increases for Dacron upon addition of
TABLE 1 Detergency of Fabrics Tested by Various
Formulations with and Without Gemini Surfactant
(% Reflectance)
Tested fabrics
a
Sample Ingredients Cotton/polyester Dacronb
1 A+1% SRP3 14.0 9.0
2 A+1% SRP3+0.5% MBOP 34.6 7.7
3 A+1% SRP3+1% MBOP 35.7 90.5
4 A+1% SRP3+1.5% MBOP 34.8 90.5
5 A+1% QCX 40.7 35.8
6 A+1% QCX+0.5% MBOP 54.5 91.2
7 A+1% QCX+1% MBOP 53.8 91.4
8 Base cationics+SRP3 1.7 5.1
Applications of gemini surfactants 299
1% MBOP. This gemini also increases very much the effectiveness of the heavier soil-
release polymer QCX (samples 5–7) and is more effective than with SRP3. Further
improvement can be realized by optimizing the polymer/ MBOP ratio. Samples 8 and 9
show the effect of MBOP on conventional cationic surfactants used as fabric softeners
and that are well known for good detergency. Here, the gemini makes the cationics
perform with high detergency and oily soil removal.
A patent [10] claimed anionic gemini surfactants of structure: [R1—N= R2—Y]2R3
with R1=R4—B—R5 (R4=hydrophobic group, B=amide group, R5=alkylene group),
R2=alkylene or its hydroxyl-substituted derivative, R3=alkylene or alkylaryl spacer, and
Y=SO3−Na+ or COO− Na+. These surfactants are mild and biologically compatible in an
ecologically sensitive environment as, for example, the N,N′-bis(2-lauramidoethyl)
ethylenediamine-N,N′-di(sodium propionate), referred to as GAn-2. Table 2 lists the
values of the Draves wetting time for GAn-2. Two conventional commercial surfactants
(MIRANOL H2M-SF and RHODAPEX ESY) blended with GAn-2 are seen to provide a
strong improvement in wetting
TABLE 2 Drave Wetting Times for Anionic
Surfactants and Their Mixtures with the Anionic
Gemini Surfactant GAn-2
Mixture of GAn-2 and
Mole fraction of GAn-2 MIRANOL H2M-SF RHODAPEX ESY
100/0 >300 >300
75/25 129 100
50/50 127 94
25/75 133 41
0/100 >300 13
1
Note: GAn-2: N,N′-bis(2-lauramidoethyl)ethylenediamine-N,N -di(sodium propionate);
RHODAPEX ESY-sodium laurylethersulfate (1 EO); MIRANOL H2M-SF-salt-free disodium
lauroamphodipropionate.
Source: Ref. 10.
ability with respect to any of the pure surfactants, indicating synergism in wetting. Thus,
the use of gemini surfactants in combination with conventional surfactants can improve
Gemini surfactants 300
the performance for blends even at a low gemini concentration. This is very desirable for
both economic and environmental reasons.
A patent [11] claimed anionic gemini surfactants such as [(C8H17C6H3O-
(C2H4O)7SO3−Na+]2CH2. This mild and environmental gemini surfactant has a good
wetting ability with Draves wetting times of 5.7 s at 0.1 % concentration. It can be used
as the essential hydrotrope component in formulations.
Cationic gemini surfactants are also disclosed along with their use in combination with
soil-release polymers in formulations that exhibit superior cleaning efficiencies when
incorporated in laundry and other cleaning detergent systems [12].
Many nonionic gemini surfactants have been developed [13–16]. For instance, the
surfactants Z(EO)a(PO)bOC(O)CR1R2R3CR1R2OC-(EO)a (PO)bZ are effective emulsifiers
for generating oil-in-water emulsions that provide improved detergency even at low
concentration (a and b are the numbers of ethylene oxide (EO) and propylene oxide (PO)
units; R1 and R2 are alkyl chains from methyl to C22, R3=D1R4D1 with R4=alkyl chain or
aryl group or C(O), and D1 is O, S, SS, SO2) [14].
An important property of nonionic gemini surfactants is their ability to lower the
concentration of free ionic surfactant in anionic-nonionic or cationic-nonionic surfactant
mixtures. Anionic conventional surfactants are known to be responsible for some toxicity
and skin irritancy in personal care formulations and for the deactivation of enzymes in
detergent systems, as they easily interact with enzymes. The addition of nonionic gemini
surfactants in small amount (mole fraction<0.1) to conventional anionic surfactants can
dramatically reduce the concentration of free anionic surfactant, by more than one order
of magnitude, resulting in the elimination of any detrimental effects or providing
performance enhancement for the surfactant-polymerenzyme mixtures. Other nonionic
gemini surfactants useful as cleaning agents are: ethoxylated-(20 EO)hydroquinone-bis-
lauryl-alcohol, ethoxylated ethylene dithio-bis-laurylalcohol, thio-bis-ethoxylate,
piperazine-coupled ethoxylate, and so forth [14]. Nonionic gemini surfactants with three
hydrophilic head groups and two lipophilic tails exhibit excellent surface-active
functionality and superior cleaning power when combined with conventional single-chain
surfactants [15,16].
Amphoteric gemini surfactants are effective mild multifunctional surfactants.
Although their cost is higher than for commercial conventional surfactants, they could be
used in small amounts in mixed surfactants for a specific requirement. For instance, the
amphoteric surfactant GAn-2 discussed earlier can be used in hard-surface cleaners,
laundry and dish detergents, liquid detergents and other washing products, and cosmetic
products, as it is extremely mild, environmentally safe, and nonirritating to both eyes and
skin [17]. This compound blended with a conventional anionic surfactant such as sodium
laurylethersulfate at a 25/75 molar ratio brings about an enhancement of surface activity
(cmc, C20) by one to two orders of magnitude greater than for a mixture of the
corresponding conventional surfactants [17].
In conclusion, if gemini surfactants can be produced at a market price not much higher
than the corresponding commercial surfactants or used at a much lower percentage than
conventional surfactants, they will have a cost-effective advantage in cleaning.
Applications of gemini surfactants 301
III.
COSMETICS AND PERSONAL CARE PRODUCTS
the natural skin lipids by mimicking the bilayer structure of the skin. Concentrates can be
used as building block for hair and skin cleaning [21]. The combination of Ceralution H
and Ceralution F can offer an efficient and flexible remedy for o/w emulsion spray
formulation. Duhring-Chamber and patch in vivo dermatological tests proved the gemini
surfactants and the Ceralution H and F to be nonirritating.
Various zwitterionic anionic and cationic gemini surfactants have been reported to
inhibit skin and eye irritation. The gemini betaine [C9H19C(O)-
NHCH2CH2(CH3)N+CH2COO−]2(CH2)2 was less irritating to the eyes than its single-
chain analog [22]. Anionic gemini surfactants such as [C11-
H23CONHCH2CH2NCH2CH2COO−Na+]2CH2CH2 showed good mildness to skin and are
environmentally safe [23]. The dicationic surfactant [C12H25N+(CH3)2CH2CONH]2(CH2)4
was nonirritating and provided anti-microbial resistance to wool fabrics [24].
The dicationic surfactant [C18H37(CH3)2N+Cl−]2CH2CHOHCH2 was reported to exhibit
hair conditioning properties superior to those of stear-alkonium chloride. Rubine dye test
showed its higher substantivity to hair, as a result of its double charge [25].
IV.
TEXTILE DYEING
The addition of surfactants to dye baths affects the dyeing parameters such as dye bath
stability, levelness of dyeing, dyeing rate, and dye uptake because of the dispersing,
emulsifying, and solubilizing ability of the surfactant. Gemini surfactants have been
found to be more effective than conventional surfactants.
Choi et al. studied the disperse dyeing of nylon-6 [26] and polyester [27] by the dye
1,4-diaminoanthraquinone (1,4-DAA) in the presence of the cationic geminis 12–3–12
and 12–6–12 and compared the results with those for the corresponding conventional
surfactant dodecyltrimethylammonium bromide (DTAB). With both fabrics, the dyeing
rate increased in the order DTAB < 12–3–12<12–6–12. The dye uptake altered slightly
with the surfactant concentration below the cmc, but it decreased hyperbolically with
increasing surfactant concentration above the cmc. This effect arises because the
surfactant micelles solubilize an increasing amount of dye and the micelle-solubilized
dye takes little part in dyeing the fiber. A linear relationship was obtained between the
amount of dye bound by the fiber and that in the dye bath, enabling the determination of
an apparent partition coefficient K. At surfactant concentration above the cmc K
increased in the order DTAB<12–6–12< 12–3–12<<water. These results indicate that K
is related to the solubilization capacity of the surfactant. An improvement in dye uptake
occurred in the presence of geminis, with a maximum dye uptake both below and above
the cmc that was 10–30% larger than with the corresponding conventional surfactant for
Nylon-6. The improvement of dye intake was somewhat lower with polyester. These two
studies were complemented by an investigation of the solubilization of various dyes by
the same surfactants [28]. As expected, the solubilization capacity was found to be
strongly dependent on the structure of the surfactant and of the dye. The authors
concluded that gemini surfactants can be used to control dyeing kinetics or to improve
dye uptake in the disperse dyeing of polyester or Nylon-6 [26,27].
Applications of gemini surfactants 303
V.
CHEMICAL INDUSTRY
Reported studies suggest that gemini surfactants can be used to replace conventional
surfactants in several sectors of the chemical industry.
A.
Polymerization
The polymerization of styrene in ternary o/w microemulsions based on the cationic
gemini surfactants [C12H25(CH3)2N+,Br−]2(CH2)s, referred to as 12-s-12 was investigated
[29]. The surfactant aggregation properties and the interfacial spontaneous curvature
could be easily tuned by changing the spacer carbon number s. This change also
influenced the formation of microemulsions. The polymerization rate in the
microemulsions was very fast because of the highly dispersed structure and the restricted
reaction space. Ultrasmall particles in the size range 5–100 nm were prepared by adding a
cross-linker to styrene. The optimum mircoemulsion formulation leading to both small
particle size and high molecular weight was found for a gemini with a spacer carbon
number s around 6 (Fig. 2a). This value constitutes a compromise between the surfactant
ability to stabilize the parent microemulsion droplets and its ability to support the
polymerization process by a high diffusive exchange mobility. It has been emphasized in
Chapter 7, Section IV that for 12-s-12 surfactants, the value s=6 is that where several
properties, the cmc for instance, go through a minimum or a maximum.
Later, the cationic gemini surfactants 12-EOx-12 [where the (CH2)s spacer of 12-s-12
surfactants is replaced by a hydrophilic and flexible spacer CH2CH2(OCH2CH2)x with
x=1–5], were used for the polymerization of styrene in styrene-water-12-EOx-12 o/w
microemulsions [30]. The single-phase region of the ternary system was the largest for
x=1 and decreased with increasing x. The micelle and microdroplet dimensions also
decreased with increasing x. Polymerization induced by 60Co γ-rays led to spherical
polystyrene particles in the nanometer size range. The particle size and molecular weight
of polystyrene were maximum at medium spacer chain length
B.
Stabilization of Latex
The zwitterionic geminis C10H21PO4−(CH2)2N+(CH3)2C12H25 and C14H29-
− +
PO4 (CH2)2N (CH3)2C8H17 have been evaluated as stabilizers of polystyrene latex [31].
They are more efficient than the commonly used combination of anionic and nonionic
surfactants in terms of providing heat and salt tolerance to the latex. Experimental tests
showed that these zwitterionic geminis give best properties when the two hydrocarbon
tails have approximately equal length. The high efficiency in latex stabilization is related
to the tendency of the gemini to form a closely packed monolayer at the particle surface.
Applications of gemini surfactants 305
C.
Paper-Sizing Dispersions, Coatings, and Paints
Cellulose-reactive and cellulose-nonreactive sizes are widely used for sizing paper
(“sizing” means covering paper with a gluelike material or sizing material; after sizing
the paper is stronger and has a smooth luster) [32]. They are used as aqueous dispersions
containing a soluble dispersing agent so that they can be easily handled in the aqueous
paper-making environment, because they are most frequently water insoluble. In general,
conventional surfactants are not used as dispersing agents for paper size dispersion
because they exhibit an antisizing effect. However, gemini surfactants have been found to
be unexpectedly effective for paper sizing, even when used at very low level [32]. The
aqueous dispersions comprise the paper-sizing compounds and water-soluble gemini
dispersants such as N-tallow pentamethylpropanediammonium chloride, or Dowfax
emulsifiers, or di- or polyquaternary ammonium chloride surfactants, such as
(C18H37(CH3)2N+[CH2CHOHCH2 (CH3)N+(C18H37)]n−1 CH3•nCl−. Thedispersions can be
used along with other paper-making ingredients, and added to the pulp slurry in the wet
end of the paper-making process, followed by formation of the sheet and drying. They
may also be used in surface sizing, where they are applied to the surface of the paper
from a size press after the sheet is formed and at least partially dried [32].
Gemini surfactants can also be used in polymer dispersions for making protective
coatings and paints, as surfactants still play a major role in almost every paint and coating
formulation. Many companies concentrate on building more effective environmentally
referred technologies, such as water-borne formulations, but conventional surfactants
limit the performance of waterborne polymers when compared to solvent-borne
formulations. The gemini surfactants EnviroGem AE and EnviroGem AD01 [33] are
readily biodegradable, low-foam wetting agents synthesized from naturally occurring
acids. They are commonly used in food and beverage flavoring and are prepared using a
clean process technology that has been extended beyond acetylenic diols. They are also
recommended in waterborne systems in automotive industry, industrial maintenance and
wood coatings, overnight varnishes, paints, and printing inks.
D.
Corrosion Inhibition
Gemini surfactants m-2-m have been investigated as potential inhibitors of corrosion of
iron in 1 M HCl by different techniques [34–37]. The results showed that the gemini
surfactants act mainly as cathodic inhibitors by adsorbing on the electrode surface and
forming a protective layer. The results also indicated that the added surfactants do not
change the proton reduction mechanism and that the inhibition efficiency increases with
the number of carbon atoms in the alkyl chain (m), with increasing surfactant
concentration and is a maximum near cmc (see Fig. 3).
In the anodic range, the inhibition efficiency also increased with increasing
concentration. An adsorbed layer of gemini molecules was stable within the
Gemini surfactants 306
VI.
PREPARATIVE CHEMISTRY
A.
Micellar Catalysis
In 1971, Bunton et al. [39] used micelles of the cationic gemini surfactants 16–4–16 and
16–6–16 to catalyze hydrolysis reactions. These surfactants were shown to be twofold to
fivefold better catalysts than the conventional surfactant CTAB (cetyltrimethylammonim
Applications of gemini surfactants 307
bromide). In addition, the catalytic efficiency was observed at a lower gemini surfactant
concentration than with CTAB. Micellar solutions of the dimeric surfactants 16-s-16
were used to modify the rate of various types of chemical reaction: nucleophilic
substitutions [39], decarboxylation [40], and cyclization [41].
Bis-quaternary ammonium surfactants could be used as phase transfer catalysts in
alkylation reactions and gave a better extractive activity than compounds generally used
for this purpose [42].
B.
Capillary Chromatography
Peptide ergot alkaloids, extracted from ergot alkaloid, are hydrogenated to form
pharmaceutically important compounds, the dihydroergotoxines. In this process, they also
undergo acid-catalyzed rearrangement to form pharmacologically inactive acid alkaloids
[43]. For Food and Drug Administration (FDA) approval, synthetic ergot alkaloids must
be tested for impurity content and stability. The impurity content was difficult to measure
by high-performance liquid chromatography or capillary electrophoresis. The micellar
electrokinetic capillary chromatography (MECC) method using cationic gemini
surfactants permitted the separation of 17 dihydroergotoxines, acid alkaloids, and
oxidation products in less than 8 min [44]. Both the didodecyl and ditetradecyl
quarternary ammonium gemini surfactants could be used, but the MECC separation could
not be achieved using conventional cationic surfactants with a C14 or C16 chain. The
separation by MECC is affected by the type and concentration of the surfactant, the pH
and ionic strength of the buffer, the background electrolyte, the type and concentration of
any organic additive, and the temperature.
C.
Preparation of Anisotropic Gold Particles
Rodlike gold particles can be obtained by using surfactants that form rodlike micelles
which then play the role of a soft template. Cationic m-s-m gemini surfactants with a
short spacer can form long threadlike micelles at a much lower concentration than
conventional surfactants. Esumi et al. [45,46] reported that anisotropic gold particles
could be prepared in a 2 wt% solution of 12–2–12 dichloride by ultraviolet photolytic
reduction of HAuCl4 added to the solution. Fibrous gold particles were obtained and their
length increased with increasing concentration of both HAuCl4 and surfactant.
In other reports [47,48], the nonionic gemini surfactant Surfynol 465 was used to
prepare colloidal silver and gold. The yield in silver particles was up to 10 times larger
than when using other methods.
D.
Gelators of Organic Solvents and Water
The self-assembly of small molecules in solution into very elongated aggregates is
necessary for the formation of gels. The self-assembling depends on the presence of
chiral centers, hydrogen-bonding, or face-to-face π–π aromatic stacking. The formation
Gemini surfactants 308
of wormlike micelles and tubules is easy in aqueous solutions of geminis with a short
spacer (see Chapter 7, Section IV.B). This also allows gemini surfactants to gel organic
solvents if they assemble into aggregates similar to those formed in water. The gemini
surfactants 16–2– 16, with L- and D-tartrate counterions can gel organic solvents
containing traces of water [49]. Gelling is already effective with a surfactant
concentration as low as 10 mM. Transmission electron microscopy revealed the presence
of long entangled helical fibers, right handed for L-tartrate, and left-handed for D-tartrate
counterions. These surfactants can gel chlorinated solvents such as CH2Cl2, CHCl3, and
Cl2CHCHCl2. For instance, one molecule of 16–2–16,L-tartrate gels about 1200 CHCl3
molecules. The gels remained unchanged after several months and turned into solutions
that were completely transparent and fluid above 40 °C but translucent and viscous upon
standing for several hours at 25°C. Chirality and hydrogen-bonding play a role in gel
cohesion. No gel formed when using an equimolar mixture of L- and D-tartrate
surfactant. In anhydrous solvents, the L- and D-tartrate surfactants tend to crystallize
instead of forming gels. A multilayered structure has been proposed for the aggregates of
16–2–16,D-tartrate in water and in organic solvents [50].
Gelation of water was reported to occur with bis-urea dicarboxylic acid gemini
surfactants [51]. Scanning electron microscopy also revealed the presence of entangled
fibers.
E.
Synthesis of Mesostructured Silica
In two separate studies [52,53], m-s-m gemini surfactants have been shown to be superior
to conventional cationic surfactants for the preparation of mesostructured silica. The pore
size was found to increase with the alkyl chain carbon number m [53]. The spacer carbon
number s affected the type of mesostructure, with the hexagonal-phase MCM-41 favored
by short spacers and the cubic-phase MCM-48 obtained with surfactants with s=10–12
[53].
F.
Environmental Surfactants
The quantity of gemini surfactant used for a given application being much lower than that
of conventional surfactants, this reduces the load of waste-water treatment. Moreover,
environmental surfactants, such as bis(sulfonate) or bis(carboxylate) gemini surfactants
containing C=C groups in the hydrophobic moiety or spacer are cleaved by ozone (see
Chapter 2, Section III.B). Their degradability is much higher than for conventional
surfactants such as LAS, SDS, for instance [54,55].
Applications of gemini surfactants 309
VII.
PHARMACEUTICAL AND BIOLOGICAL APPLICATIONS
A.
Antimicrobial Activity
Dicationic m-s-m surfactants with a short spacer show high antimicrobial activity, up to
100 times larger than the commonly used germicides dodecylbenzyldimethylammonium
bromide or 2-ethoxycarbonylpentadecyltri-methylammonium bromide [56,57]. It has
been reported [58] that that the minimum inhibitory concentration (MIC) that
characterizes the antimicrobial activity of the gemini surfactants m-4-m for
Staphylococcus aureus, Escherichia coli, and Candida albicans is a maximum for m=11–
12, after which it decreases (Fig. 4).
The dicationic gemini surfactants [C12H25N+(CH3)2CH2CONH]2Y •2Cl− [with
Y=(CH2)4 or (CH2)2SS(CH2)2] were found to be more effective against many micro-
organisms other than CTAB [24]. Arginine-based gemini surfactants (see Chapter 2,
Scheme 7) have been shown to possess a broad range of antimicrobial activity, with the
Gram-negative bacteria more resistant than Gram-positive bacteria (see Table 3). They
also have a low toxicity [59,60].
B.
Perturbation of the Human Erythrocyte
Dicationic gemini surfactants m-4-m, with m=8, 11, 13, and 16, can induce shape
alteration, vesiculation, hemolysis, and phophatidylserine exposure in
C.
Gene Therapy and Gene Transfection
Gene therapy (i.e., the replacement or addition of gene) is an important topic in life
science. It depends on the effective techniques for the safe introduction of the selected
gene into living cells. Recently, gemini surfactants have been found to be particularly
promising as potential vehicles for the transport of bioactive molecules. The European
Network on Gemini Surfactants [62] has produced 250 new gemini surfactants that are
efficient gene transfection agents for the safe delivery of functional genes into cells and
then into the nucleus. Camilleri et al. [63–66] reported the synthesis of five peptide-based
Applications of gemini surfactants 311
cationic gemini surfactants (see two examples in Fig. 5) having the ability to transfer
plasmid DNA containing the luciferase gene into cells. These surfactants are easy-to-
handle solids that are readily soluble in aqueous media over a wide range of pH. They
include a thioether linkage
unsaturated chains have the efficiency of transfecting Chinese hamster ovary cell line
(CHO-K1) [68].
Other cationic gemini surfactants based on tartaric acid appended with biocompatible
palmitoyl tails and with lysine or the combined lysine-ethylenediamine head groups have
been reported [69]. They show activity in luciferase gene-transfection assay, but with a
considerable toxicity.
VIII.
MISCELLANEOUS
IX.
CONCLUSIONS
Because of the enormous variety in the structure of gemini surfactants, their superior
interfacial activity, and highly effective performance, dimeric or trimeric surfactants have
the potential of being used in household cleaning agents, personal care products,
cosmetics, emulsifiers, wetting agents, solubilizers, food and drug additives, textile
dyeing and finishing agents, antimicrobial agents, polymerization, inhibition of corrosion,
Applications of gemini surfactants 313
REFERENCES
68. Fielden, M.L.; Perrin, C.; Kremer, A.; Bergsma, M.; Stuart, M.C.; Camilleri, P.; Engberts,
J.B.F.N. Eur. J. Biochem. 2001, 268, 1269.
69. Buijnsters, P.; Rodriguez, C.L.G.; Willighagen, E.L.; Sommerdijk, N.; Kremer, A.; Camilleri,
P.; Feiters, M.C.; Nolte, R.J.M.; Zwanenburg, B. Eur. J. Org. Chem. 2002, 8, 1397–1406.
70. Chen, X.; Wang, J.; Shen, N.; Luo, Y.; Li, L.; Liu, M.; Thomas, R.K. Langmuir 2002, 18, 6222.
71. Jaeger, D.A.; Wang, Y.; Pennnigton, R.L. Langmuir 2002, 18, 9259.
72. Eastoe, J.; Dominguez, M.S.; Wyatt, P.; Beeby, A.; Heenan, R.K. Langmuir 2002, 18, 7837.
73. Karthaus, O.; Shimomura, M.; Hioki, M.; Tahara, R.; Nakamura, H. J. Am. Chem. Soc. 1996,
118, 9174.
74. Gallardo, B.S.; Abbott, N.I. Langmuir 1997, 13, 203.
75. Yan, X.; Janout, V.; Regen, S.L. Macromolecules 2002, 35, 8243.
Index
Activity coefficient, 39
Adsorbed layer structure, 85
Adsorption
barriers, 80
capacity, 67
dynamics, 295
equilibrium constant, 67
isotherms, 67
Adsorption of gemini surfactants
at air/water interface, 65
on clay, 88
on graphite, 85
on mica, 85
oil/water interface, 70
on silica, 84
solid/water interface, 83, 85
on titanium oxide, 88
on wool fibers, 88
Aggregate, (see Micelle)
Alkyl chains
interactions in gemini surfactants, 103
self-coiling, 104
Amino-acid based gemini surfactants, 254, 269–275
aggregate shape, 164
aggregation of, 269–275
antimicrobial activity, 270, 273, 275, 315
gene-transfer vectors, 270–273
Krafft temperature, 269
Amphoteric gemini surfactants, (see Zwitterionic gemini surfactants)
Antimicrobial activity and properties, 9, 88, 261, 263, 269, 270, 273, 275, 308, 315
Antiredeposition, 303
Applications of gemini surfactants, (see also Antimicrobial activity and properties)
capillary chromatography, 313
care products, 306, 307
cleaning agents and detergents, 302–306
corrosion inhibition, 311
cosmetics, 306, 307
deodorizing formulations, 306
emulsification, 305, 307
enzyme deactivation, 305
gelation of organic solvents, 313
gene therapy, 316 (see Transfection)
micellar catalysis, 312
Index 317
Bilayer adsorption, 85
Bilayer from gemini surfactants, 38, 49, 55, 59, 59, 61, 156, 163, 230
Bilayer stacks, 220, 225
Binary mixture, (see Phase behavior)
Biodegradation of gemini surfactants, 296
Bjerrum length, 44
Bolaform surfactants
micelle size, 148
model, 54
structure, 2
Branched (connected) cylindrical micelles, 38, 58–60, 62, 159, 199, 202
Branching density, 202
Branching in trimeric surfactant micelles, 202
Debye length, 44
Dendrimers
interactions with gemini, 133
Dermatological test, 308
Diffusion controlled
adsorption, 79, 296
Index 319
Fluorescence probing
micelle micropolarity, 167
micelle microviscosity, 169
Fluorinated gemini surfactants
aggregation, 268–269
antimicrobial activity, 269
vesicle formation, 268
Foam
decay time, 81
stability mechanisms, 82
stabilizers, 81
Foaming, 69, 80–83, 302, 311
Free energy of micelle growth, 149
Free energy of micellization
experimental results, 126
Index 320
Krafft Temperature
amino-acid based surfactants, 269
anionic gemini surfactants, 110
carbohydrate based surfactants, 258
cationic gemini surfactants, 38, 55, 111
effect of the hydrophilic head group on, 282–284
sugar gemini surfactants, 115
Oligomeric surfactants
branched cylindrical micelles from, 159
closed ring micelles from, 159
Index 323
cmc, 124–125
micelle ionization degree, 144
micelle micropolarity, 168–169
micelle microviscosity, 169–171
micelle shape, 161
micelle size, 151
rheology of solutions of, 185
synthesis of cationic, 15, 18
Optical reflectometry, 69
Oxidation sensitive gemini surfactants, 318
Packing
constraints, 42
optimal, 40, 42, 50, 61
Packing parameter, 42–43,
micelle growth and, 148–149
micelle shape and, 55, 159, 216
micelle size and, 151
phase behavior and, 59, 216
Paints, 310, 311
Paper-sizing dispersions, 310
Performances of gemini surfactants
ability to lower surface tension, 285, 289, 293
cloud point, 282
effect of the hydrophilic head group on, 282
effect of the hydrophobic group on, 288
effect of the spacer group on, 293
foaming, 286, 291, 294
Krafft point, 282, 288
lime-soap dispersing ability, 286, 291, 294
pollutant removal, 292
solubility in water, 282, 288
stability to calcium, 286,
stain removal, 288
Peptide-based gemini, (see Amino-acid based gemini surfactants)
Personal care products, 281, 306, 307
Phase
coagel, 212
cubic bicontinuous, 212
cubic micellar, 212, 217
cylindrical, (see Hexagonal phase)
gel, 212
hexagonal, 212, 218
lamellar, 212, 218
sequence, 216, 217
Phase behavior
comparison of monomeric and gemini surfactant, 223
of arginine-derived gemini surfactants, 219
effect of additive on, 225
effect of alkyl chain length and nature, 220–222
effect of oil and cosurfactant, 226
Index 324
Vesicles from gemini, 156, 163–164, 258, 260, 261, 263, 268, 230
Vesicles from surfactant mixtures, 245, 248–250
Viscoelasticity
of branched polymers, 192
of carbohydrate-based gemini surfactants, 259
of neutral linear polymers, 191
of polyelectrolytes, 191
of wormlike micelles, 189
Viscoelastic relaxation by
chemorheological processes, 194
end interchange, 194
reptation, 191, 192
reptation/reversible scission, 192
Viscosity, 4
Index 328
Zero-shear viscosity
concentration dependence of, 197
Zwitterionic gemini surfactants 28–30, 230, 254, 265–268, 306
patents, 266–268
rheological behavior of, 266
sponge-forming, 265
stabilization of latex by, 266
vesicle formation by, 265–266