Design of dialkyl surfactants from nitrilotriacetic acid as head group

Article abstract of DOI:10.1039/c3ra45543b

New double-chain surfactants can be designed in which nitrilotriacetic acid plays the role of glycerol in phospholipids. The synthetic methodology is applied to the synthesis of (2-(bis(2-dodecylamino)-2-oxoethyl)amino)acetic acid. This is a symmetric dimeric surfactant but the methodology allows the synthesis of asymmetric double-chain surfactants as well. The Krafft temperature is 16-18 °C. The critical aggregation concentration of the new surfactant (4.76 × 10^-5 mol dm^-3 in NaOH 0.01 M and 25 °C; surface tension measurements) is two hundred times lower than that for sodium 2-dodecanamidoacetate, which can be considered its monoalkyl surfactant analog, indicating the enhanced surfactant properties resulting from the presence of two hydrophobic alkyl chains. The shape of the aggregates was investigated by transmission electron microscopy. At 40 °C, the main structure present in solution corresponds to vesicles and, from the size distribution of their radii, values of (0.34 ± 0.06)kT and 56.7 ± 2.6 nm were obtained for the effective bending elasticity constant and the spontaneous radius of curvature of vesicles, respectively. The Royal Society of Chemistry 2014.

Full text of DOI:10.1039/c3ra45543b

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Design of dialkyl surfactants from nitrilotriacetic
acid as head group
Cite this: RSC Adv., 2014, 4, 6869
a
Juan V. Trillo, Francisco Meijide, a Jose Vazquez Tato,a Aida Jover,a Victor H. Soto,b
*
´
´
Santiago de Frutosa and Luciano Galantinic
New double-chain surfactants can be designed in which nitrilotriacetic acid plays the role of glycerol in
phospholipids. The synthetic methodology is applied to the synthesis of (2-(bis(2-dodecylamino)-2-
oxoethyl)amino)acetic acid. This is a symmetric dimeric surfactant but the methodology allows the
synthesis of asymmetric double-chain surfactants as well. The Krafft temperature is 16–18 ꢀC. The
critical aggregation concentration of the new surfactant (4.76 ꢁ 10ꢂ5 mol dmꢂ3 in NaOH 0.01 M and
25 ꢀC; surface tension measurements) is two hundred times lower than that for sodium 2-
dodecanamidoacetate, which can be considered its monoalkyl surfactant analog, indicating the
enhanced surfactant properties resulting from the presence of two hydrophobic alkyl chains. The shape
of the aggregates was investigated by transmission electron microscopy. At 40 ꢀC, the main structure
present in solution corresponds to vesicles and, from the size distribution of their radii, values of (0.34 ꢃ
0.06)kT and 56.7 ꢃ 2.6 nm were obtained for the effective bending elasticity constant and the
spontaneous radius of curvature of vesicles, respectively.
Received 3rd October 2013
Accepted 18th December 2013
DOI: 10.1039/c3ra45543b
www.rsc.org/advances
phospholipid in animal and plants, and it is a key building
block of membrane bilayers. Generally, as it is well-known,
Introduction
During the last few years dimeric surfactants (gemini,1 hybrid2
and phospholipids) have attracted considerable interest
because they are characterized by critical micelle concentra-
tions that are one to two orders of magnitude lower than those
for the corresponding monomeric surfactants. Dimeric surfac-
tants are also much more efficient than monomeric in
decreasing the surface tension of water.1 Some zwitterionic
gemini surfactants, with phosphate and ammonium as ionic
groups,3 exhibit extraordinary properties and structures form-
ing gels, micelles, vesicles, and coacervates. In the coacervate
region, someone resembles sponges. As a result of the screening
of many surfactant derivatives, it is possible to predict con-
dently which compounds will self-assemble into a sponge
morphology.
Phosphatidylcholine is a fat derivative of glycerol carrying a
saturated hydrophobic palmitic acid, an unsaturated hydro-
phobic oleic acid, and a phosphate group forming an ester with
choline. Consequently a zwitterionic surfactant is formed. We
have to notice the different nature of the two fatty chains.
Phosphatidylcholine is usually the most abundant
these phospholipids have a strong tendency to self-aggregate in
water forming bilayers as in vesicles. Length and nature of
linked chains can be modied to obtain different families of
compounds. Two common examples are dipalmitoylphospha-
tidylcholine (DPPC, the major constituent of the pulmonary
surfactant) and dioleoylphosphatidylcholine (DOPC). It has
been observed a slight decrease in the average area per lipid as
well as a linear increase in the bilayer thickness with increasing
chain lengths.4 Positive correlation between most probable
vesicle size and acyl chain lengths have also been noticed.5
Many other variants are possible as well, sphingomyelins being
known examples.
On the other hand, a series of glycerol-based double- or
triple-chain surfactants with two ionic (either sulfonate, sulfate
or carboxylate) groups have been prepared by Zhu et al.6 The
sulfate and carboxylate derivatives have higher water solubil-
ities than the corresponding sulfonate ones, and the triple-
chain surfactants show excellent surface-active properties, such
as micelle forming and ability to lower the surface tension,
compared with the corresponding single- and double-chain
surfactants.
The structure of nitrilotriacetic acid 1 (Fig. 1) resembles the
structure of glycerol since both compounds present a central
atom from which three organic functions emerge. In glycerol
the central atom is a carbon one and hydroxy groups are the
organic functions, while a nitrogen atom and carboxylic acids
are the characteristics of 1.
a
´
´
Departamento de Quımica Fısica, Facultad de Ciencias, Universidad de Santiago de
Compostela, Avda. Alfonso X El Sabio s/n, 27002 Lugo, Spain
b
´
´
´
´
´
Escuela de Quımica, Centro de Investigacion en Electroquımica y Energıa Quımica
´
(CELEQ), Universidad de Costa Rica, San Jose, Costa Rica
c
`
Dipartimento di Chimica, Universita di Roma “Sapienza”, P. le A. Moro 5, 00185
Roma, Italy
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 6869–6877 | 6869

View Article Online
RSC Advances
Paper
from the knowledge of structure and properties of natural
phospholipids.
Results and discussion
The synthesis of 12-ida-12 was performed according to the three
steps procedure outlined in Scheme 1. Amide 4 (R1 ¼ C12H25
)
7
´
was prepared as described by Puigmartı-Luis et al. The ida
function 5 was introduced by means of nucleophilic substitu-
tion of the a-chloride with iminodiacetic acid according to the
method published by van Wyk et al.8 In situ closure of the imi-
nodiacetic acid 5 to the anhydride with 3-ethyl-1-[3-(dimethy-
Fig. 1 General structure of surfactants 2 from nitrilotriacetic acid 1.
Structure of (2-(bis(2-dodecylamino)-2-oxoethyl)amino)acetate (12- lamino)propyl]carbodiimide hydrochloride (EDCI), followed by
treatment with the R2NH2 amine (R2 ¼ R1)9 afforded the
ida-12, 3).
diamide 12-ida-12 (3).
The 12-ida-12 surfactant was characterized by measurement
Having all previous information in mind, new derivatives
can be designed in which nitrilotriacetic acid plays the core role
of glycerol in phospholipids. With appropriate substituents,
new surfactants with structure 2 may be obtained. For that
of its Kra temperature, Tk,10 and critical aggregation concen-
tration, cac, as the aggregation of a surfactant has to be studied
above its Tk and its cac. The measurements were performed in
NaOH (0.01 M, pH ¼ 12) aqueous solutions. This high pH value,
purpose, in 2, at least one of the R groups (say R1) must have a
which ensures complete ionization of the carboxylic acid, was
hydrophobic nature (as, for instance, a long alkyl chain). The
nature of the second R group (say R2) allows to obtain
chosen to prevent possible effects observed for some carboxylic
surfactant systems at lower pH values. Although there should
not be much difference in carboxyl group deprotonation at
symmetric (R1 ¼ R2) or asymmetric (R1 s R2) surfactants which
could emulate the characteristics of phospholipids mentioned
basic pH values, it is known that myristic acid crystallizes as
above. However, as we will comment on below, this structure
1 : 1 mixture with its sodium salt at pH 9.8 and as pure sodium
analysis does not mean that the synthesis of the new surfactants
salt upon addition of sodium chloride, but gives long bers
starts from 1.
(curds) of a carboxylate monohydrate at pH 11.5. According to
In this paper, we probe the correctness of this approach for
Trager et al.11 the curvature of the myristate bilayer is enforced
¨
designing new surfactants, starting with the symmetric
compound (2-(bis(2-dodecylamino)-2-oxoethyl)amino)acetate
(compound 3 with R1 ¼ R2 ¼ C12H25). We will use the acronym
12-ida-12 (“ida” corresponds to IminoDiAcetic acid used in the
synthesis). The behavior of this compound in water has been
studied by means of surface tension, uorescence of pyrene as a
probe, transmission electron microscopy (TEM), and light
scattering techniques. The compound presented here is an
example of molecular biomimetics as its design is approached
by hydration, not by repulsive forces between anions. It is also
well known the effect of pH on the aggregation behavior of some
bile salts in water, even forming gels at pH values close to
neutrality.12 In these systems hydrogen ions have the tendency
to be adsorbed onto bile salt micelles as they form,13 and acidic
species in micelles and at the interface are present in a more
appreciable extent than in the bulk aqueous solution.14 Thus
micelles of natural bile salts are mixed micelles formed by
species with a different degree of ionization.14,15 It seems that
Scheme 1 Synthesis of surfactants with the general structure 3. Descriptions correspond to the synthesis of 12-ida-12, where R1 ¼ R2 ¼ C12H25
.
DMF: dimethylformamide; DMAP: N,N-dimethyl-4-aminopyridine; EDCI: 3-ethyl-1-[3-(dimethylamino)propyl]carbodiimide hydrochloride.
Reaction yields are also indicated.
6870 | RSC Adv., 2014, 4, 6869–6877
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
hydrogen ions are adsorbed more weakly onto smaller aggre- breaking point of g vs. ln[12-ida-12] plot, the critical aggregation
gates than onto larger ones.13
concentration and the surface tension at cac, gcac, were deter-
For measuring Tk, a solution of the surfactant (2.5 ꢁ mined. Values are recompiled in Table 1.
10ꢂ4 mol dmꢂ3) was slowly cooled (approx. at 1 ꢀC hꢂ1) till 3 ꢀC
The value obtained for cac at 25 ꢀC (temperature at which
and then heated (at the same rate). During the cooling process a most values have been published) is approximately two
sharp change in the scattered light16 (accompanied by a visual hundred times lower than the cmc of the N-dodecyl derivative of
clarity/translucence change)17 was recorded at 16 ꢀC. When glycine (2-dodecanamidoacetate) for which cmc values of 8.3 ꢁ
heating the translucent/clear change was observed at 18 ꢀC. 10ꢂ3–0.014 mol dmꢂ3 have been published.18,19 This surfactant
This narrow temperature range is taken as Tk.
has an amide group as linkage between the hydrophobic and
Surface tension measurements were carried out at 25 ꢀC, and hydrophilic moieties and is probably the closer monoalkyl
aer observing the effect of temperature and ageing of samples surfactant analog of 12-ida-12 found in the literature. The
on the type of aggregates (see below TEM images) such comparison is also advantageous with sodium laurate, for
ꢀ
measurements were also performed at 40 C.
which cmc values higher than those for 2-dodecanamidoacetate
Fig. 2 shows the dependence of the surface tension of the have been published by Mori and Matsumura and others.20–24
solution with the surfactant concentration. The typical prole The comparison between these three surfactants suggests that
for a surfactant in aqueous solution is observed and from the the lower cac value for 12-ida-12 may be ascribed to the presence
of the two alkyl chains and not to the presence of the amide
groups.
The cac value is comparable to those obtained by Zhu et al.6
for the series of glycerol-based double- or triple-chain surfac-
tants with two ionic (sulfonate, sulfate, and carboxylate) groups,
and by Yoshimura and Esumi25 for a ring-type trimeric surfac-
tant (3C10NAc-Tz) of 1,3,5-tris[1-(carboxyalkyl)amino]triazine
with three lauryl chains. Other anionic trimeric surfactants
show even lower cmc values. All of them are recompiled in Table
1. However the comparison is not straightforward as in these
trimeric surfactants the ratio [number of alkyl chains/number
of ionic groups] is one, but it is two for 12-ida-12. The reduction
of cmc of trimeric surfactants respect to the corresponding
monomeric and dimeric surfactants has been also found for
cationic triquaternary ammonium surfactants by Zana et al.27
Guo et al.28 have obtained three trimeric quaternary ammonium
surfactants having adamantane cores with cmc values in the
range 0.5–2 ꢁ 10ꢂ3 mol dmꢂ3
.
Fig. 2 Dependence of the surface tension of water solutions (NaOH
0.01 M) of 12-ida-12 with its concentration at 25 ꢀC.
Table 1 Adsorption and aggregation parameters of 12-ida-12, sodium 2-dodecanamidoacetate, sodium laurate, glycerol derivatives, 3C10NAc-
Tz and 3CntaAm (see text)
2
˚
Compound
12-ida-12
Temp. (ꢀC) cac (mol dmꢂ3
)
gcac (mN mꢂ1
)
106 G (mol mꢂ2
)
ao (A ) Ref.
25
40
4.76 ꢁ 10ꢂ5
5.2 ꢁ 10ꢂ5
3.3 ꢁ 10ꢂ5
40.2
31.7
32.1
3.36
4.49
5.81
49
This work
Freshly prepared
Aer 4 days
37
29
Sodium 2-dodecanamidoacetate
Sodium laurate
25
25
8.3 ꢁ 10ꢂ3–0.014 39
63
18 and 19
0.016–0.026
8.1 ꢁ 10ꢂ5
1.4 ꢁ 10ꢂ5
9 ꢁ 10ꢂ6
36.8
46.9
20–24
R ¼ C10H21; R0 ¼ CH3, Z ¼ (CH2)3SO3Na
36.0
28.0
27.0
29.0
R ¼ R0 ¼ C10H21, Z ¼ (CH2)3SO3Na
R ¼ R0 ¼ C10H21, Z ¼ SO3Na
R ¼ R0 ¼ C10H21, Z ¼ CH2CO2Na
4.0 ꢁ 10ꢂ5
20
25
6
3C10NAc-Tz
3C8taAm
3C10taAm
3C12taAm
1.1 ꢁ 10ꢂ5
9.2 ꢁ 10ꢂ7
8.34 ꢁ 10ꢂ6
6.41 ꢁ 10ꢂ6
28.6
39.9
33.3
36.9
4.84
1.84
4.88
4.42
34
91
34
38
25
26
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 6869–6877 | 6871

View Article Online
RSC Advances
Paper
Fig. 3 TEM images of a 9.2 ꢁ 10ꢂ4 mol dmꢂ3 surfactant solution
(in NaOH 0.01 M) for freshly prepared samples (top) and after ageing
for four days at 25 ꢀC.
TEM images were collected at 25 ꢀC in freshly prepared
samples and aer ageing for several days. Fig. 3 shows two
typical TEM images at initial time and aer four days.
These images conrm the existence of an aggregation
process as small dark objects, having some tendency to asso-
ciate in larger ones, are evident. However, no differences can be
noticed between samples of different ageing time.
However, examples in which the temperature effect on
aggregation processes and on the morphology of the aggregates
is highly important are well known.29,30 In this system,
Fig. 4 TEM images of a 9.2 ꢁ 10ꢂ4 mol dmꢂ3 surfactant solution (in
NaOH 0.01 M) at initial time (top) and after four days (center and
bottom) at 40 ꢀC. Vesicles: A and B (see text); open vesicle: C; asso-
ciated particles: D.
increasing the temperature to 40 ꢀC induced a different growth (pyrene) at 40 C. The dependence of the ratio of intensities of
process, and vesicle like structures are evident (A in Fig. 4), the rst (373 nm) and third (384 nm) vibronic peaks of pyrene,
some of them having a large size (B). At this temperature we I1/I3, is related to the polarity of the microenvironment where
have noticed that the number of vesicle structures increases pyrene is located. The dependence of I1/I3 with the concentra-
with time, but no signicative differences are observed aer tion of the surfactant is plotted in Fig. 6. A typical sigmoid
four days. Some open vesicles (C) and associated particles (D) prole is obtained with an inection point at 4.64 ꢁ 10ꢂ5 mol
ꢀ
are also observed.
dmꢂ3. This value corresponds to the cac of the surfactant and
Consequently, surface tension measurements were also compares well with the value obtained by surface tension
carried at 40 ꢀC at zero time and aer four days (Fig. 5). The cac measurements in freshly prepared solutions. At high surfactant
is more clearly dened aer four days and a decrease in its value concentrations I1/I3 reaches a plateau value around to 1.0. This
(from 5.2 ꢁ 10ꢂ5 to 3.3 ꢁ 10ꢂ5 mol dmꢂ3) is also observed. This limiting value is lower than those for sodium laurate (I1/I3 z
may be due to a different arrangement of the molecules at the 1.1)21 and for 3C10NAc-Tz (I1/I3 z 1.2).25 For vesicles, values
interface as the ao values suggest (see below).
ranging from 0.9 (ref. 31 and 32) to 1.67 (ref. 33) have been
The cac for 12-ida-12 was also estimated by means of steady published. It is accepted that in these aggregates pyrene is
state uorescence measurements in the presence of a probe solubilized in the lipophilic core of the bilayer.34 Furthermore,
6872 | RSC Adv., 2014, 4, 6869–6877
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
the very unpolar cyclohexane),35 and the sodium salts of p-tert-
butylphenyl amide of cholic and ursodeoxycholic acids which
form tubes (I1/I3 ¼ 1.19 and 1.1, respectively).30,36 A third
example corresponds to N-(2-hydroxydodecyl)-L-alanine which
forms bilayer structures (I1/I3 ¼ 1.65–1.27) and N-(n-dodecyl)-L-
alanine which form micelles (I1/I3
mentioned above for 12-ida-12 is compatible with less closed
structures than micelles, as vesicles.
¼
1.059). The value
The uorescence spectrum of pyrene (Fig. 6) also shows a
broad band around 480 nm, generally attributed to the uo-
rescence emission of excimers,35,37 as the energy of the excited
state of excimers is lower than for an excited monomer. Exci-
mers are formed when an excited-state monomer interacts with
a close enough ground-state monomer forming an excited
dimer. In this band the vibrational structure observed in
monomers is lost. The low intensity of this band does not
change with time.
Fig. 5 Dependence of the surface tension of water solutions (NaOH
0.01 M) of 12-ida-12 with its concentration at 40 ꢀC for freshly
prepared samples (,) and after ageing them for four days (-).
The surface behavior may be analyzed in terms of the
adsorption Gibbs equation, which relates the change in the
equilibrium surface tension with changes in the chemical
potentials of all of the solutes at constant temperature. From
this isotherm, the surface excess concentration, G, is deter-
mined by eqn (1).
ꢀ
ꢁ
1
vg
G ¼ ꢂ
(1)
nRT vln½12-ida-12ꢄ
where R is the universal gas constant, T is the absolute
temperature, and n is given by38
2
n
þ
n ¼ n þ
(2)
ꢂ
n þ n scs=c
þ
þ
This equation shows that the prefactor n in the Gibbs
equation depends on the concentration and stoichiometric
coefficients of the surfactant (c, nꢂ, and n+) and also on both the
s
stoichiometry (n+ ) and the concentration, cs, of an added inert
electrolyte. When the concentration of electrolyte is in a large
excess with respect to the surfactant concentration, the pre-
factor becomes n ¼ nꢂ that is, n ¼ 1. Once G is known, the
surface area per molecule can be determined by the eqn (3),
ꢂ
ꢃ
1020
2
˚
ao
A
¼
(3)
GNA
Fig. 6 Top: Dependence of the I1/I3 fluorescence intensities ratio of
pyrene solubilized in freshly prepared water solutions (NaOH 0.01 M)
of 12-ida-12 with its concentration at 40 ꢀC. The pyrene concentration
was 1.49 ꢁ 10ꢂ5 mol dmꢂ3. Bottom: Fluorescence spectrum of
pyrene in a solution of 12-ida-12 1.00 ꢁ 10ꢂ3 mol dmꢂ3 (NaOH 0.01 M)
at 40 ꢀC.
where NA is Avogadro's number. The experimental conditions
used imply cs [ c, and therefore n ¼ 1. The values obtained for
G and ao at both temperatures are shown in Table 1.
2
ꢀ
˚
The value obtained at 25 C for ao (¼49 A ) is close to the
value determined for sodium laurate by Mori and Matsumura20
and lower than the one for sodium 2-dodecanamidoacetate.19
Menger et al.39 have recently questioned the validity of the Gibbs
equation to determine the surface area per molecule at the
air/water interface. According to this criticism, the Gibbs anal-
ysis overestimates the area per molecule at saturation. However
such an overestimation is hardly applicable to the present case.
The cross section of one hydrocarbon chain,40 Ac z 20 A
matches well with the area occupied by each molecule of long
alkyl chain alcohols in the monolayer. For instance, for 1octa-
decanol a value of 23 ꢃ 0.1 A per molecule has been
signicative differences in I1/I3 values are observed for different
structures (as micelles, vesicles or tubes) originated by the same
or similar surfactants. One example is the formation of micelles
by lauryl sulfonate betaine (LSB) (I1/I3 ¼ 1.19) and bis(2-ethyl-
hexyl) sulfosuccinate (AOT) (I1/I3 ¼ 0.83), and vesicles formed by
a mixture LSB–AOT: 3/7 (I1/I3 ¼ 0.96).31 Another example is the
case of sodium deoxycholate35 which forms small aggregates at
high pH (I1/I3 ¼ 0.70, a value which is very close to the one for
2
˚
2
˚
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 6869–6877 | 6873

View Article Online
RSC Advances
Paper
measured.41 This is an expected agreement because of the
alcohol molecules are highly ordered at the interface. For a
R2
Ro
(
"
#)
ꢀ
ꢁ
2
2
8p
Ro
R
CR
¼
CR exp
ꢂ
kB 1 ꢂ
(4)
o
kT
double tail surfactant the cross section of two alkyl chains will
2
correspondingly be 40–42 A .42,43 Such a value is a minimum one
˚
In this equation, CR and CR are the molar concentrations of
for ao because of steric constraints between the alkyl chains of
adjacent molecules in the monolayer. The small difference
between the experimental ao and 2Ac may be due to the elec-
trostatic ion–ion repulsions of ionic headgroups. This agree-
ment suggests that a monolayer at the air/water interface is
formed.
o
aggregates of size R and Ro, respectively, Ro is the spontaneous
radius of curvature of vesicles, kB is the effective bending elas-
ticity constant, and kT is the thermal energy.
By tting this equation to data of Fig. 6 by a non-linear
optimization procedure, the effective bending elasticity
constant of vesicles and the spontaneous radius of curvature
can be deduced. Values of kB ¼ (0.34 ꢃ 0.06)kT and Ro ¼ 56.7 ꢃ
2.6 nm are obtained. Fig. 7 shows a good result for the t.
kB is very close to the one obtained for vesicles formed by of a
p-tertbutylphenyl amide derivative of cholic acid, (0.36 ꢃ 0.06)
kT,30 and for cetyltrimethyl-ammonium bromide/sodium octyl
sulfonate catanionic vesicles, (0.7 ꢃ 0.2)kT.54 However, most of
the published values for bilayer vesicles of lipids are within the
range (1–20)kT.53,55–58 Under a low polydispersity condition, eqn
(4) can be simplied by expanding its right side around Ro,
allowing a quick estimation of Ro and kB from the average radius
of vesicles and its standard deviation (see Denkov et al.53 for
details). This represents an extraordinary simplicity for the
analysis of experimental results since only the knowledge of the
normal distributions of the vesicle radius is required. However,
a consequence of the low value for the effective bending
constant is that the vesicles have a rather broad size distribution
and this simplied analysis is not advisable. Since kB < kBT the
vesicles are stabilized by thermal uctuations (rather than by
the spontaneous curvature), and their size distribution is rather
broad.59
Above its cmc, sodium laurate forms spherical micelles with
radii lower than 2 nm (fully compatible with the length of the
alkyl chain) and an average aggregation number N ¼ 89.24 The
aggregation number of 3C10NAc-Tz at the cmc was found to be
about N ¼ 580 ꢃ 20,25 while from dynamic light scattering, DLS,
measurements, two main populations with hydrodynamic
diameters of 41 ꢃ 9 and 338 ꢃ 167 nm were observed at 4.7 ꢁ
10ꢂ5 mol dmꢂ3 of 3C10-NAc-Tz. For this compound Yoshimura
and Esumi accept that the trimeric surfactants are more likely to
form micelles in the bulk solutions. The aggregation numbers
of 3C8taAm, 3C10taAm, and 3C12taAm at the cmc were about 10,
82, and 63, respectively.26
For monoalkyl ionic surfactants as sodium laurate and
2
oleate,20–24,44 values around 47 A have been published while for
˚
2
45–47
˚
phospholipids the accepted values are in the range 60–67 A .
These values are consistent with the surface area per lipid head
2
˚
group in egg phosphatidylcholine vesicles as values of 74 A and
2
48
˚
61 A are obtained for outer and inner lipids, respectively.
The ao value observed at 40 ꢀC in freshly prepared samples is
slightly lower than 2Ac and, in our opinion, previous comments
are still applicable. As cac, ao is time dependent and aer four
2
˚
days the value is 29 A , not far from Ac. This probably means
that a surfactant bilayer is now formed at the interface. This
interpretation would be in agreement with Rosen et al.49 who
have proposed the formation of multilayer structures to explain
low ao values measured for some gemini surfactants. Tsubone
et al.50 have also accepted the formation of multilayer lms.
Similar comments would probably apply to 3C10NAc-Tz and to
3CntaAm (with three carboxylate headgroups and three hydro-
carbon chains; n ¼ 10, 12) since the geometry and charge of
their head group would suggest a higher value for ao than the
observed one.26
A recount of the size of the vesicles (A type; total number ¼
ꢀ
ꢀ
231) observed at 40 C gives a mean value for the radius of R ¼
48 ꢃ 12 nm, and the distribution of Fig. 7, where f is the fraction
of vesicles with sizes in a given interval.
By applying the thermodynamic theory of aggregation from
Israelachvili51 and the Helfrich elasticity theory for
membranes,52 Denkov et al.53 have deduced eqn (4) for vesicles,
which represents the size distribution of the vesicles in the
sample.
ꢀ
From the values of R for the vesicles and the surface area per
molecule ao, the aggregation number of vesicles formed by
12-ida-12 can be roughly estimated as 1.8 ꢁ 104.
The size of the aggregates in aqueous solution was also
investigated by DLS measurements. The results shown in Fig. 8
correspond to a sample with a concentration 6.0 ꢁ 10ꢂ4 mol
dmꢂ3 aer four days, and shows the volume weighted distri-
bution aer analysis of the intensity correlation function by the
NLLS method at a scattering angle of 90ꢀ. This analysis gives a
more meaningful representation of size polydispersity than
cumulant methods.60
Fig. 7 Histogram corresponding to the radii distribution of vesicles of
the surfactant 12-ida-12 in the experimental conditions of Fig. 4. f is
the fraction of vesicles with sizes in a given interval. The superimposed
continuous line is the best fit of eqn (4) to it.
The most signicant populations are in the ranges 81.6–
164.8 and 333.1–673.2 nm, with maxima at 97.2 and 397.1 nm,
respectively, for the hydrodynamic diameter of particles.
6874 | RSC Adv., 2014, 4, 6869–6877
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
temperature for 5 h. The ethanol was removed in vacuo and the
aqueous phase was extracted with ether (3 ꢁ 20 ml). The pH of
the aqueous phase was adjusted to 2 with aqueous HCl to
precipitate the product. The solid product 5 was washed with
water till neutral pH and nally dried in vacuo at 50 ꢀC over-
night. Yield: 75%.
1H NMR (300 MHz, d6-DMSO, d/ppm): 7.87 (t, J ¼ 5.8 Hz, 1H,
–NH); 3.42 (s, 4H, –CH2COOH); 3.28 (s, 2H, -NCH2CO); 3.08
(q,
J
¼
6.9 Hz, 2H, –NHCH2CH2); 1.39–1.16 (m, 20H,
–NHCH2(CH2)10CH3); 0.85 (t, J ¼ 6.5 Hz, 3H, –CH3). 13C NMR (75
MHz d6-DMSO, d/ppm): 173.1 (COOH); 171.0 (CONH); 58.9 (N–
CH2CO–); 56.3 (N–CH2–COOH); 39.1 (–NHCH2CH2–); 31.9; 29.6;
29.3; 29.3; 27.0; 22.7; 14.5 (–(CH2)10CH3).
Fig. 8 Size distribution of aggregates from dynamic light scatteringꢂo4f
12-ida-12 at 40 ꢀC after four days at a concentration of 6.0 ꢁ 10
mol dmꢂ3 in aqueous solution and [NaOH] ¼ 0.01 M.
Step 3:9 A solution of 5 (0.75 mmol, 0.27 g) in dried DMF
(8 ml) was treated with EDCI, (0.79 mmol, 0.15 g) at 25 ꢀC. The
mixture was stirred at room temperature under N2 for 1 h before
dodecylamine (0.79 mmol, 0.15 g) was added, and the solution
was stirred at for 24 h at 40 ꢀC. The reaction mixture was poured
into 10% aqueous HCl (50 ml). The ltrate was washed with
10% HCl (50 ml) and with water till neutral pH. The solid was
Dimensions of the lower population compares favorably with
diameters from TEM images of type A vesicles. The larger size
population, which represents a lower proportion, can corre-
spond to large type B vesicles, open vesicles (C) and associated
particles (D). The limits of the accuracy of DLS analysis have
been commented on by Viseu et al.60
ꢀ
dried in vacuo at 45 C and recrystallized from hot ethanol. A
white solid was obtained (3, 12-ida-12). Yield: 50%.
Conclusions
1H NMR (300 MHz, d6-DMSO, d/ppm): 8.08 (t, 2H J ¼ 5.8 Hz,
–NH–); 3.36 (s, 2H, –CH2COOH); 3.23 (s, 4H, –NCH2CO–); 3.05
(q, 4H, J ¼ 6.9 Hz, –NHCH2CH2–); 1.39–1.24 (m, 40H,
NHCH2(CH2)10CH3); 0.85 (t, 6H, J ¼ 6.6 Hz, –CH3). 13C NMR
(75 MHz d6-DMSO, d/ppm): 173.1 (COOH); 170.7 (–CONH–);
58.9 (–COCH2N); 56.4 (–NCH2–COOH); 39.1 (–CH2CH2NH–);
31.9; 29.8; 29.7; 29.6; 29.4; 29.3; 27.0; 22.7; 14.4 (–(CH2)10CH3).
It has been proved that new surfactants can be designed from
nitrilotriacetic acid, playing a similar core role to glycerol in
phospholipids. The proposed synthetic route allows the
synthesis of symmetric or asymmetric dialkyl chain surfactants.
The methodology is illustrated with the compound 12-ida-12.
For its synthesis dodecyl amine has been used as the carrier of
the alkyl chains. TEM images strongly suggest that above a
critical aggregation concentration, the surfactant form vesicles.
FT-IR (n/cmꢂ1): 3256.7, 3334.6 (NH and OH), 2912.9, 2847.5
~
(CH), 1697.0 (C]O, acid), 1663.4 (C]O, amide), 1566.8 (NH,
amide).
Mass spectrum (ESI-FIA-TOF): m/z calculated for C30H59N3O4
525.81, found 526.45. Elemental analysis: Calculated for
Experimental
Synthesis
C
30H59N3O4: N¼ 7.99%, C ¼ 68.53%, H ¼ 11.31%; found: N ¼
Step 1:7 Chloroacetyl chloride (55 mmol, 4.4 ml) and N,N- 8.30%, C ¼ 68.08%, H ¼ 11.40%. M.p. 91.2–92.0 C.
ꢀ
dimethyl-4-aminopyridine, DMAP, (0.1 mg) were dissolved in
dried CH2Cl2 (50 ml) under N2 atmosphere. The mixture was
cooled to 0 C and dodecylamine (50 mmol, 9.3 g) with trieth-
Equipment
ꢀ
ylamine (8.4 ml) in dried CH2Cl2 (10 ml) was slowly (1 h) added Surface tension measurements were carried out in a KRUSS
dropwise. The resulting suspension was stirred for 3 h at room model K-10-ST tensiometer equipped with a Wilhelmy plate.
temperature. The solvent residue was ltered and the ltrate Temperature was kept constant by recirculating water from a
was washed with CH2Cl2. The organic phase was washed with PolyScience 9100 thermostat. The Wilhelmy plate method was
aqueous HCl (1 M, 3 ꢁ 15 ml). The resulting organic phase was used. Solutions were prepared according to the step-by-step
dried with MgSO4, ltered and concentrated in vacuo. The crude dilution-extraction method,61 with a total volume of 10 ml and
product was puried by silica gel column chromatography dilution steps of 3 ml. A complete set of measurements was
eluting with ethyl acetate : hexane (6 : 4). The resulting white performed in less than two hours. In order to keep a pH constant
ꢀ
solid 4 was dried in vacuo at 50 C overnight. Yield: 65%.
value, the dilution solvent was a 0.01 M NaOH aqueous solution.
1H NMR (300 MHz, d6-DMSO, d/ppm): 6.63 (s, 1H, NH); 3.97 Temperature was kept constant either at 25 ꢀC or 40 ꢀC.
(s, 2H, ClCH2–); 3.23 (q, J ¼ 6.9 Hz, 2H, NHCH2–); 1.54–1.27
Steady-state uorescence measurements, using pyrene
(Sigma) as a probe, were recorded on a Cary Eclipse (excitation
(m, 20H, –NHCH2(CH2)10CH3); 0.83 (t, J ¼ 6.4 Hz, 3H, –CH3). 13
C
NMR (75 MHz d6-DMSO, d/ppm): 163.4 (CONH–); 43.5 (CH2Cl); wavelength 336 nm, excitation slit 2.5 nm, emission slit 2.5 nm).
39.8 (–NHCH2–); 31.9; 29.3; 29.2; 29.3; 29.0; 28.9; 26.5; 22.8; 14.6 Solutions containing pyrene were prepared according to the
(–(CH2)10CH3).
method described by Zana and Guveli.62 Temperature was kept
Step 2:8 4 (R1 ¼ C12H25) (5.85 mmol, 1.45 g) and iminodi- constant at 40 C.
ꢀ
acetic acid (11.7 mmol, 1.55 g) in 50% aqueous ethanol (80 ml)
TEM images were obtained at room temperature in a JEOL
at pH ¼ 11 (adjusted with NaOH solution) were kept at reux JEM-1011, operated at 80 kV, equipped with a MegaView III
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 6869–6877 | 6875

View Article Online
RSC Advances
Paper
camera. For sample preparation, a drop of the solution (kept at
40 ꢀC for most experiments) was deposited onto a carbon-coated
copper grid and le for a few minutes to allow complete grid
permeation. Aer that, the grid was deposited on lter paper to
absorb the drop excess and measured immediately.
6 Y.-P. Zhu, A. Masuyama, Y. Kirito, M. Okahara and
M. J. Rosen, J. Am. Oil Chem. Soc., 1992, 69, 626–632.
´
´
´
7 J. Puigmartı-Luis, A. Minoia, A. Perez del Pino, G. Ujaque,
´
C. Rovira, A. Lledos, R. Lazzaroni and D. B. Amabilino,
Chem. – Eur. J., 2006, 12, 9161–9175.
A Brookhaven instrument constituted by a BI-2030AT digital
correlator with 136 channels and a BI-200SM goniometer was
used for dynamic (DLS) light scattering measurements. The
light source was a Melles Grio He–Ne laser operating at 632 nm.
In the DLS experiments the intensity–intensity autocorrela-
8 A. J. Van Wyk, P. J. Fourie, P. Bekker, S. Wood, K. Smith,
T. Gerber and J. Du Preez, Curr. Appl. Radiopharmacol.,
Proc. Int. Symp. Radiopharmacol., 4th, 1986, 27–34.
9 S. Cheng, D. D. Comer, J. P. Williams, P. L. Myers and
D. L. Boger, J. Am. Chem. Soc., 1996, 118, 2567–2573.
tion function was measured at a particular q value, and related 10 D. H. Everett, Manual of Symbols and Terminology for
to the normalized electric eld autocorrelation function g1(q,s)
Physicochemical
Quantities
and
Units,
IUPAC,
by the Siegert relation. The autocorrelation function was
BUTTERWORTHS, LONDON, 1971.
¨
¨
analyzed by a NLLS algorithm (from Brookhaven soware). 11 O. Trager, S. Sowade, C. Bottcher and J.-H. Fuhrhop, J. Am.
From the so called apparent diffusion coefficient, Dapp, an Chem. Soc., 1997, 119, 9120–9124.
apparent hydrodynamic radius Rapp can be calculated by the 12 A. Jover, F. Meijide, E. Rodrıguez Nunez and J. Vazquez Tato,
Stokes-Einstein equation, Rapp ¼ kBT/6phDapp, where kB is the Langmuir, 2002, 18, 987–991.
Boltzmann constant, T is the absolute temperature and h is 13 M. Swanson-Vethamuthu, M. Almgren, P. Hansson and
the solvent viscosity.63
J. Zhao, Langmuir, 1996, 12, 2186–2189.
´
˜
´
Dust was eliminated by ltering the samples through Albet 14 M. C. Carey, Hepatology, 1984, 4, 66S.
PTFE lters with a pore size of 0.45 mm at room temperature, the 15 H. Igimi and M. C. Carey, J. Lipid Res., 1980, 21, 72.
¨
volume of the sample being about 5.0 ml. The temperature of 16 R. Klein, M. Kellermeier, D. Touraud, E. Muller and W. Kunz,
the samples was kept constant within 0.5 ꢀC by a circulating
J. Colloid Interface Sci., 2013, 392, 274–280.
water bath. The samples were placed in the cell for at least 17 D. A. Jaeger, Y. Wang and R. L. Pennington, Langmuir, 2002,
30 min prior to measurement to allow for thermal equilibration. 18, 9259–9266.
NMR experiments were performed in a Varian Mercury 18 W. M. Lineld, J. Am. Oil Chem. Soc., 1978, 55, 87–92.
300 instrument. The solvent used was d6-DMSO.
19 R. Bordes, J. Tropsch and K. Holmberg, J. Colloid Interface
Sci., 2009, 338, 529–536.
FT-IR spectrum was performed in an ABB Bomen model MB
102 instrument. Solid powdered samples were placed on a 20 K. Mori and S. Matsumura, J. Oleo Sci., 2012, 61, 609–620.
GOLDEN-GATE accessory for solid samples.
21 R. Eising, A. C. Felippe and J. B. Domingos, J. Dispersion Sci.
Technol., 2012, 33, 75–82.
Mass spectra were performed in a Bruker Microtof equip-
ment. A Flash EA TCD Thermo Finnigan elemental model 1112 22 P. S. Sales, R. H. de Rossi and M. A. Fernandez, Chemosphere,
analyzer with thermal conductivity detector was used for 2011, 84, 1700–1707.
elemental analysis of 12-ida-12. The melting points were 23 R. Eising, S. Mores, I. C. Bellettini, A. C. Felippe, A. G. Dal-Bo
determined using a Gallenkamp capillary melting point appa-
ratus (model MPD 350 BM 2.5; SANYO Gallenkamp PLC).
OriginPro 7 (OriginLab Corporation) was used for tting all
experimental data to the appropriate equations.
and D. Zanette, Quim. Nova, 2008, 31, 2065–2070.
˜
24 A. Rodriguez-Pulido, A. Casado, M. Munoz-Ubeda,
E. Junquera and E. Aicart, Langmuir, 2010, 26, 9378–9385.
25 T. Yoshimura and K. Esumi, Langmuir, 2003, 19, 3535–
3538.
26 T. Yoshimura, N. Kimura, E. Onitsuka, H. Shosenji and
K. Esumi, J. Surfactants Deterg., 2004, 7, 67–74.
Acknowledgements
´
The authors thank the Ministerio de Ciencia y Tecnologıa, 27 R. Zana, H. P. Levy, D. Papoutsi and G. Beinert, Langmuir,
Spain (Project MAT2010-19440) for nancial support. J. V. Trillo
1995, 11, 3694–3698.
also thanks for a scholarship.
28 H. Zhu, J. Guo, C. Yang, S. Liu, Y. Cui and X. Zhong, Synth.
Commun., 2013, 43, 1161–1167.
¨
29 S. Polarz, C. Bahrle, S. Landsmann and A. Klaiber, Angew.
Notes and references
Chem., Int. Ed., 2013, 52, 1–7.
´
1 R. Zana, Adv. Colloid Interface Sci., 2002, 97, 205–253.
30 V. H. Soto Tellini, A. Jover, F. Meijide, J. Vazquez Tato,
´
2 M. Alvarez Alcalde, A. Jover, F. Meijide, L. Galantini,
L. Galantini and N. V. Pavel, Adv. Mater., 2007, 19, 1752–
1756.
´
´
N. V. Pavel, A. Antelo and J. Vazquez Tato, Langmuir, 2009,
25, 9037–9044.
31 L. Zhai, J. Zhang, Q. Shi, W. Chen and M. Zhao, J. Colloid
Interface Sci., 2005, 284, 698–703.
3 F. M. Menger, Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 4819–
4822.
32 L. Zhai, X. Lu, W. Chen, C. Hu and L. Zheng, Colloids Surf., A,
2004, 236, 1–5.
¨
¨
4 P. S. Niemela, M. T. Hyvonen and I. Vattulainen, Biophys. J.,
2006, 90, 851–863.
33 S. Roy and J. Dey, Langmuir, 2005, 21, 10362–10369.
5 T. J. Racey, M. A. Singer, L. Finegold and P. Rochon, Chem. 34 E. Brueckner, P. Sonntag and H. Rehage, Langmuir, 2001, 17,
Phys. Lipids, 1989, 49, 271–288.
2308–2311.
6876 | RSC Adv., 2014, 4, 6869–6877
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
´
˜
´
´
35 A. Jover, F. Meijide, E. Rodrıguez Nunez, J. Vazquez Tato, 49 M. J. Rosen, J. H. Mathias and L. Davenport, Langmuir, 1999,
´
M. Mosquera and F. Rodrıguez Prieto, Langmuir, 1996, 12,
15, 7340–7346.
1789–1793.
50 K. Tsubone, T. Ogawa and K. Mimura, J. Surfactants Deterg.,
2003, 6, 39–46.
36 F. Meijide, J. V. Trillo, S. de Frutos, L. Galantini, N. V. Pavel,
´
V. H. Soto, A. Jover and J. Vazquez Tato, Steroids, 2012, 77, 51 J. Israelachvili, Intermolecular and Surface Forces, Santa
1205–1211. Barbara, 3rd edn, 2011
37 D. Sahoo, V. Narayanaswami, C. M. Kay and R. O. Ryan, 52 W. Helfrich, Z. Naturforsch. C, 1973, 28, 693–703.
Biochemistry, 2000, 39, 6594–6601.
53 N. D. Denkov, H. Yoshimura, T. Kouyama, J. Walz and
38 A. J. Prosser and E. I. Franses, Colloids Surf., A, 2001, 178, 1–40.
K. Nagayama, Biophys. J., 1998, 74, 1409–1420.
39 F. M. Menger, L. Shi and S. A. A. Rizvi, J. Am. Chem. Soc., 54 H. T. Jung, B. Coldren, J. A. Zasadzinski, D. J. Iampietro and
2009, 131, 10380–10381.
E. W. Kaler, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 1353–
40 K. Okuyama, Y. Soboi, N. Iijima, K. Hirabayashi, T. Kunitake
1357.
and T. Kajiyama, Bull. Chem. Soc. Jpn., 1988, 61, 1485–1490. 55 E. Evans, V. Heinrich, F. Ludwig and W. Rawicz, Biophys. J.,
41 E. L. Prime, D. N. H. Tran, M. Plazzer, D. Sunartio, 2003, 85, 2342–2350.
A. H. M. Leung, G. Yiapanis, S. Baoukina, I. Yarovsky, 56 W. Rawicz, K. C. Olbrich, T. McIntosh, D. Needham and
G. G. Qiao and D. H. Solomon, Colloids Surf., A, 2012, 415,
47–58.
E. Evans, Biophys. J., 2000, 79, 328–339.
57 S. Dieluweit, A. Csiszar, W. Rubner, J. Fleischhauer,
S. Houben and R. Merkel, Langmuir, 2010, 26, 11041–11049.
58 G. Popescu, T. Ikeda, K. Goda, C. A. Best-Popescu,
M. Laposata, S. Manley, R. R. Dasari, K. Badizadegan and
M. S. Feld, Phys. Rev. Lett., 2006, 97, 218101.
42 C. Tanford, The Hydrophobic Effect: Formation of Micelles and
Biological Membranes, Wiley, New York, 1980.
43 R. Nagarajan, Langmuir, 2002, 18, 31–38.
44 W. D. Harkins and H. Zollman, J. Am. Chem. Soc., 1926, 48,
69–80.
59 B. Coldren, R. Van Zanten, M. J. Mackel, J. A. Zasadzinski
and H.-T. Jung, Langmuir, 2003, 19, 5632–5639.
45 N. Kucerka, J. F. Nagle, J. N. Sachs, S. E. Feller, J. Pencer,
A. Jackson and J. Katsaras, Biophys. J., 2008, 95, 2356–2367. 60 M. I. Viseu, K. Edwards, C. S. Campos and S. M. B. Costa,
46 S. H. White, R. E. Jacobs and G. I. King, Biophys. J., 1987, 52,
663–665.
Langmuir, 2000, 16, 2105–2114.
61 A. Jover, F. Meijide, V. Mosquera and J. Vazquez Tato,
´
47 N. Kucerka, J. Gallova, D. Uhrikova, P. Balgavy, M. Bulacu,
J. Chem. Educ., 1990, 67, 530.
¨
S.-J. Marrink and J. Katsaras, Biophys. J., 2009, 97, 1926– 62 R. Zana and D. Guveli, J. Phys. Chem., 1985, 89, 1687–1690.
1932.
63 An Introduction to Dynamic Light Scattering by
Macromolecules, ed. K. S. Schmitz, Academic Press, Boston,
1990.
48 C. Huang and J. T. Mason, Proc. Natl. Acad. Sci. U. S. A., 1978,
75, 308–310.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 6869–6877 | 6877